摘要
淬火、控制冷却的冷却速度和目标温度是关系到钢板性能的关键因素 ,但由于冷却过程的水量、水温、辊道速度等对冷却目标温度的影响属多变量非线性过程 ,用简单的解析模型通常精度较低 ,难以达到精确控制的要求。采用人工神经网络技术对无约束淬火机目标温度进行快速预报 ,结果表明 ,预测精度高 。
This the cooling speed and goal temperature of hardening and conttroling cool are key factors of effect plate property.In cooling process,it is nonlinear and multivariable process that water quantity,water temperatur and roller speed have an effect on goal temperatur.The precision of simple model of controling cool is lower ,it is difficulty to the demand of precise control. The make use of neural network technology that forecast goal temperature for hardening machine without restraint With the result that shown:forecast precision high,quick and accurate.
出处
《冶金设备》
2003年第3期17-21,共5页
Metallurgical Equipment
关键词
神经网络
目标温度
快速预报
Neural network Goal temperature Quick forecast
