首页> 外国专利> Learning model generation method, database construction method, mill setup setting method, rolled material manufacturing method, processing target manufacturing method, and learning model generation device

Learning model generation method, database construction method, mill setup setting method, rolled material manufacturing method, processing target manufacturing method, and learning model generation device

机译:学习模型生成方法,数据库构建方法,轧机设置设置方法,轧制材料制造方法,加工目标制造方法和学习模型生成装置

摘要

PROBLEM TO BE SOLVED: To provide a learning model generation method and a learning model generation device capable of generating a learning model which outputs a set value of a production facility satisfying a desired operation index. SOLUTION: A learning model generation method according to the present invention obtains operation performance data of a production facility including an operation condition of the production facility, an actual value of a variable serving as an operation index of the production facility, and an actual value of a set value of the production facility. The first learning step to learn the variable learning model that inputs the operating conditions and outputs the predicted value of the variable, and the variable prediction by inputting the operating conditions included in the operating performance data into the variable learning model. A selection step that calculates a value and selects the operation performance data corresponding to the processing target based on the comparison between the predicted value of the variable and the actual value, and the operation condition is input using the selected operation performance data, and the set value The second learning step of learning the learning model of the set value which outputs the predicted value of is included. [Selection diagram] Fig. 2
机译:要解决的问题:提供一种学习模型生成方法和能够生成学习模型的学习模型生成装置,其输出满足期望操作索引的生产设施的设定值。解决方案:根据本发明的学习模型生成方法获得生产设施的操作性能数据,包括生产设施的操作条件,作为生产设施的操作索引的变量的实际值以及实际值生产设施的设定值。学习输入操作条件的可变学习模型并输出变量的预测值的第一学习步骤,以及通过将包括在运行性能数据中的操作条件输入到可变学习模型中的可变预测。基于对变量的预测值和实际值之间的比较来计算值并选择与处理目标对应的操作性能数据,并且使用所选择的操作性能数据输入操作条件,以及该组赋予包括输出预测值的设定值的学习模型的第二学习步骤。 [选择图]图2

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号