基于改进人工鱼群算法的1050铝合金变形抗力模型优化  

The deformation stress model optimization of alloy based on improved artificial fish swarm algorithm

作　　者：杨景明[1,2] 许爽[1] 赵新秋[1,2] 车海军[1,2] 

机构地区：[1]国家冷轧板带装备及工艺工程技术研究中心,河北秦皇岛066004 [2]燕山大学工业计算机控制工程河北省重点实验室,河北秦皇岛066004

出　　处：《计算机与应用化学》2013年第3期281-285,共5页Computers and Applied Chemistry

基　　金：河北省科学技术研究与发展计划基金资助项目(编号:10212157)

摘　　要：变形抗力是轧制力模型预测的重要参数,其模型的精确度直接影响轧制力的预报精度。为了得到更为精确的变形抗力数学模型,文中以据河南某铝热连轧厂1+4连轧精轧机组为研究对象,根据大量现场实测稳定段数据,应用轧制理论反推得到变形抗力回归所用数据后,将模型线性化再应用最小二乘法对1050系铝合金变形抗力进行了线性回归,得到初始参数后运用改进鱼群算法对影响变形抗力的参数进行了优化,仿真结果表明用改进鱼群算法优化出的参数得到的变形抗力模型比用最小二乘法回归出的变形抗力模型应用于轧制力预报的误差更小,所以,在进行轧制力预报时选择改进鱼群算法优化后的变形抗力模型。精确的轧制模型是轧制过程中的轧制预报的关键,对于指导生产,提高产品质量有着重要的实践意义。The deformation stress is an important parameter during the process of rolling force model prediction. The accuracy of the model directly influences the rolling schedule. In order to get a more accurate model of deformation stress, the paper took one factory of aluminum hotl＋4 finishing rolling in Henan as the investigation objective. According to lots of actual measured data, the data which model of deformation stress is needed were calculated by use of rolling theory, the stress model equation can be linearized and then the least square method was applied to the linear regression of deformation stress model of alloy 1050. The initial parameters were got and then use improved artificial fish swarm algorithm to optimize the influencing parameters of the model. The results show that the mode using improved artificial fish swarm algorithm has the better error term than the least square method. Precise rolling model is one of the most important factor in the process of rolling forecast, it has important practical significance to guide the production, improve the quality of products.

关 键 词：铝热连轧 轧制力模型 变形抗力 最小二乘法 人工鱼群算法 

分 类 号：TG339[金属学及工艺—金属压力加工]