摘要
阐述将改进的BP神经网络用于直流电弧炉水冷炉壁问题曲线的辨识。首先分析了偏弧偏向角和偏移量和水冷炉壁的温度的关系,然后通过辨识水冷炉壁的温度,根据公式确定偏弧的着弧点,计算出偏弧的偏向角和偏移量。该方法的应用为直流电弧炉偏弧控制打下坚实的基础。
BP neural network is improved and applied to the identification of the water-cooling furnace walls curves of the Direct Current Electrical Arc Furnace (DC EAF). First the relation of deflection angle and the offset of arc deflection with the temperature of the water-cooling furnace walls is analyzed, and then the arc point of arc deflection is validated by discriminating the temperature of the water-cooling furnace walls, and finally the deflection angle and the offset of the arc deflection are calculated out. It lays the foundation for the arc deflection control of DC EAF.
出处
《机电工程技术》
2005年第12期33-34,77,共3页
Mechanical & Electrical Engineering Technology
关键词
BP神经网络
直流电弧炉
系统辨识
偏弧
水冷炉壁
BP neural network
Direct Current Electrical Arc Furnace (DC EAF)
system identification
arc deflection
water-cooling furnace walls curves,