基于GA-BP神经网络的弯辊力预设定模型研究与应用
|
摘要
针对弯辊力预设定模型精度低、弯辊力调整到设定值时间长的问题,基于GA-BP神经网络模型,建立了冷连轧机弯辊力预设定优化模型。结果表明,利用GA-BP神经网络优化模型使弯辊力实际值达到预设定值的调整时间平均缩短了115 ms,提高了钢材成材率和板形质量。
In view of low precision of the preset model and a long adjusting time to reach the set value for the roller bending force,an optimized model for the roller bending force in a tandem cold rolling mill was established based on GA-BP neural network. The experimental results showed that the model optimized with GA-BP neural network has not only shortened the adjusting time to reach preset values by 115 ms on average,but also improved product yield and the quality of strip shape.
In view of low precision of the preset model and a long adjusting time to reach the set value for the roller bending force,an optimized model for the roller bending force in a tandem cold rolling mill was established based on GA-BP neural network. The experimental results showed that the model optimized with GA-BP neural network has not only shortened the adjusting time to reach preset values by 115 ms on average,but also improved product yield and the quality of strip shape.
引文
[1]王文明,钟掘,谭建平.板形控制理论与技术进展[J].矿冶工程,2001,21(4):70-72.
[2]田宝亮,牛培峰,马啸飞,等.1450冷轧机板形控制系统变形抗力补偿研究[J].矿冶工程,2014,34(1):107-109.
[3]杨景明,李永泽,王洪庆,等.铝热粗轧机自适应辊缝动态设定[J].矿冶工程,2014,34(2):108-112.
[4]孙浩,杨景明,呼子宇,等.基于数据相似性的铝热连轧轧制力模型自学习[J].矿冶工程,2015,35(3):127-130.
[5]蒋中华,陈刚,陈振华,等.AZ31镁合金轧制板材的中温变形性能研究[J].矿冶工程,2008,28(6):97-100.
[6]袁武华,杨寿智,陈振华,等.轧制加工对7075/Si Cp复合材料组织性能的影响[J].矿冶工程,2005,25(2):62-65.
[7]张翀,张新明,刘胜胆,等.轧制变形对7A55铝合金晶间腐蚀的淬火敏感性的影响[J].矿冶工程,2007,27(6):69-71.
[8]魏立新,林鹏,王利平.基于Elman网络预测的神经网络控制器在冷轧APC系统中的应用研究[J].东北大学学报,2015,35(6):130-134.
[9]杨景明,顾佳琪,闫晓莹,等.基于改进遗传算法优化BP网络的轧制力预测研究[J].矿冶工程,2015,35(1):111-115.
[10]杨景明,闫晓莹,顾佳琪,等.基于改进粒子群优化RBF神经网络的轧制力预报[J].矿冶工程,2014,34(6):110-114.
[11]John S,Sikdar S,Swamy P K,et al.Hybrid neural-GA model to Predict and Minimise Flatness Value of Hot Rolled Strips[J].Journal of Materials Processing Technology,2008,195(1-3):314-320.
[12]薛涛,杜凤山,孙静娜.基于有限元与神经网络的板形调控功效[J].钢铁,2012,47(3):56-60.
[2]田宝亮,牛培峰,马啸飞,等.1450冷轧机板形控制系统变形抗力补偿研究[J].矿冶工程,2014,34(1):107-109.
[3]杨景明,李永泽,王洪庆,等.铝热粗轧机自适应辊缝动态设定[J].矿冶工程,2014,34(2):108-112.
[4]孙浩,杨景明,呼子宇,等.基于数据相似性的铝热连轧轧制力模型自学习[J].矿冶工程,2015,35(3):127-130.
[5]蒋中华,陈刚,陈振华,等.AZ31镁合金轧制板材的中温变形性能研究[J].矿冶工程,2008,28(6):97-100.
[6]袁武华,杨寿智,陈振华,等.轧制加工对7075/Si Cp复合材料组织性能的影响[J].矿冶工程,2005,25(2):62-65.
[7]张翀,张新明,刘胜胆,等.轧制变形对7A55铝合金晶间腐蚀的淬火敏感性的影响[J].矿冶工程,2007,27(6):69-71.
[8]魏立新,林鹏,王利平.基于Elman网络预测的神经网络控制器在冷轧APC系统中的应用研究[J].东北大学学报,2015,35(6):130-134.
[9]杨景明,顾佳琪,闫晓莹,等.基于改进遗传算法优化BP网络的轧制力预测研究[J].矿冶工程,2015,35(1):111-115.
[10]杨景明,闫晓莹,顾佳琪,等.基于改进粒子群优化RBF神经网络的轧制力预报[J].矿冶工程,2014,34(6):110-114.
[11]John S,Sikdar S,Swamy P K,et al.Hybrid neural-GA model to Predict and Minimise Flatness Value of Hot Rolled Strips[J].Journal of Materials Processing Technology,2008,195(1-3):314-320.
[12]薛涛,杜凤山,孙静娜.基于有限元与神经网络的板形调控功效[J].钢铁,2012,47(3):56-60.