【24h】

Prediction of Abnormality Mould Friction on Slab Caster Based on Neural Network

机译:基于神经网络的板坯连铸机结晶器摩阻预测。

获取原文
获取原文并翻译 | 示例

摘要

Based on the analysis results of measured mould friction (MDF) of abnormality slab casting , an abnormality prediction method on MDF has been built employing neural network in combination with two models for ramp and pulse judgments. A set of software for abnormality prediction by MDF has been developed. The simulation results for on-line measured MDF data accord basically with the records of abnormality in steel plant. The software can forecast most abnormalities such as breakout, break of submerged entry nozzle, mould level fluctuation and other abnormalities, and it can make prediction several minutes before temperature system warning. The results show that the method demonstrates remarkable potential for application to prediction of abnormalities.
机译:根据异常铸坯的实测铸模摩擦力(MDF)分析结果,建立了基于神经网络结合两种模型进行斜率和脉冲判断的MDF异常预测方法。已经开发了一套用于通过MDF进行异常预测的软件。在线测量的MDF数据的模拟结果基本上与钢厂的异常记录相符。该软件可以预测大多数异常,例如破裂,浸入式喷嘴的破裂,模具高度波动和其他异常,并且可以在温度系统警告前几分钟进行预测。结果表明,该方法具有广阔的应用前景。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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