本发明公开了一种转炉后吹碳含量动态预测方法及装置,该方法包括：以当前转炉生产过程为新案例,历史转炉生产过程为历史案例,根据新案例的主吹阶段工艺参数,基于案例推理算法检索出历史案例中的相似案例及其后吹阶段工艺参数；利用相似案例的后吹阶段工艺参数对碳含量预测模型进行训练；将新案例的后吹阶段工艺参数输入训练好的碳含量预测模型,实现吹炼后期碳含量的实时动态预测。本发明针对目前转炉终点静态控制模型和动态控制模型存在的问题,基于案例推理和长短期记忆网络建立了转炉预测模型,实现了对转炉吹炼后期碳含量的实时动态

The invention discloses a dynamic prediction method and device for the carbon content in the back blowing of the converter. The method comprises: Taking the current converter production process as a new case and the historical converter production process as a historical case, according to the process parameters in the main blowing stage of the new case, the similar cases in the historical case and the process parameters in the back blowing stage are retrieved based on the case-based reasoning algorithm; The carbon content prediction model is trained by using the process parameters in the back blowing stage of similar cases; The process parameters of the post blowing stage of the new case are input into the trained carbon content prediction model to realize the real-time dynamic prediction of carbon content in the later stage of blowing. The invention provides a reference model for the prediction of the end-point of the converter based on the case-based reasoning model and the dynamic control of the end-point of the converter, which can improve the prediction rate of the end-point of the converter based on the case-based reasoning model.