摘要
根据河南某1+4铝热连轧厂的大量现场实测轧制数据,利用最小二乘法对5052系铝合金变形抗力模型进行了回归,并用细菌觅食优化群算法对变形抗力模型的参数进行了优化。最后,将细菌觅食优化算法优化后的变形抗力模型与最小二乘回归出的变形抗力模型分别应用于轧制力的预报。仿真结果表明用细菌觅食优化算法优化后的变形抗力模型应用于轧制力预报时误差更小,在进行轧制力预报时选择细菌觅食优化算法优化后的变形抗力模型。
According to lots of measured actual data from one factory of 1 ＋ 4 aluminum hot continuous rolling, the least square method was applied to the linear regression of deformation resistance model of aluminum alloy 5052 ,and then Bacterial Foraging Optimization Algorithm was used to optimize the in- fluencing parameters of the model ,finally the rolling forces were respectively predicted by BFOA and the liner regression. The results showed that the model optimized by BFOA had less error than the least square method when applying prediction rolling forces.
出处
《轻金属》
CSCD
北大核心
2013年第7期61-65,共5页
Light Metals
基金
国家冷轧板带装备及工艺工程技术研究中心开放课题资助
河北省工业计算机控制工程重点实验室开放课题资助
关键词
铝热连轧
轧制力
变形抗力模型
最小二乘法
细菌觅食优化算法
aluminum hot continuous rolling
rolling foree
deformation resistance model
least square method
Bacterial Foraging Optimization Algorithm
