作　　者： ; ; ; ; ;

机构地区： 东北大学材料与冶金学院轧制技术及连轧自动化国家重点实验室

出　　处： 《宽厚板》 2007年第1期1-5,共5页

摘　　要： 为了达到厚板生产中的强度和屈强比等性能指标,本文在用神经元网络对屈服强度和抗拉强度建模的基础上,结合粒子群优化算法对粗轧开轧温度、中间坯厚度、终轧温度、终冷温度及冷却速率等生产工艺参数进行了优化。优化结果与实验室热轧实验及工业试生产结果的对比表明,本模型能有效地优化厚板生产过程的工艺参数,从而为优化工艺或柔性化生产工艺的设计提供指导。 In the present paper, based on the model of yield strength and tensile strength established by artificial neural network combined with PSO （Particle Swarm Optimization）, the optimization of rolling process parameters such as rough rolling start temperature, temperature-holding thickness, finish rolling temperature, finish cooling temperature and cooling rate was carried out in order to obtain the desired strength and yield ratio of heavy plate. The comparison between the results of optimization and the results of hot rolling trials in laboratory and industrial trial production shows that the model is able to effectively optimize the process parameters during heavy plate production, thereby guiding the design of optimized process or more flexible production process.

关 键 词： 厚板 神经网络 粒子群算法

领　　域： [金属学及工艺]