摘要
通过分析层流冷却过程中钢板终冷温度的各种影响因素,建立了终冷温度的BP神经网络预报模型,确定了网络的输入、输出量。运用"试凑法"对不同网络结构进行了训练,最终确立了神经网络模型结构,并用所建模型对终冷温度进行预测。结果显示:网络预测值与现场实测值比较逼近,误差基本维持在±7℃,证明所建模型是合适的,可以用于生产现场的指导。
By analyzing the various factors of final cooling temperature of steel in the laminar cooling, the BP neural network prediction model of final cooling temperature was established and the input and output of the network were determined. The neural network model structure was established after the different network structures were trained by ＂trial and error＂, and the final cooling temperature was predicted by the model. The re- suits showed that the network predictive value and the measured value were approximate, and the error was maintained at ＋7℃. The model was appropriate and could be used to guide the production.
出处
《冶金丛刊》
2011年第6期16-19,共4页
Metallurgical Collections
基金
陕西省工业攻关项目支助(2009K07-20)
关键词
中厚板
BP神经网络
层流冷却
终冷温度
plate
BP neural network
laminar cooling
final cooling temperature
