摘要
为提高TWIP钢的屈服强度同时保留较好的塑性,利用BP神经网络和遗传算法对热处理工艺参数进行优化。以退火温度、保温时间和冷却方式为输入,屈服强度和伸长率的乘积为输出,建立3-4-1的BP神经网络模型,再通过遗传算法寻优,得到屈服强度和伸长率的乘积最大时TWIP钢的热处理工艺参数组合。结果表明,优化后的热处理工艺为:退火温度768℃、保温时间35 min、冷却方式为炉冷,并通过试验验证了预测结果的准确性。
In order to improve the yield strength and mean while retain the better plasticity of TWIP steel,BP neural network and genetic algorithm were used to optimize heat treatment process parameters.Taking annealing temperature,holding time and cooling method as input,the product of yield strength and elongation as output,a 3-4-1 BP neural network model was established.Through the optimization of genetic algorithm,the heat treatment process parameters with the maximum product of yield strength and elongation were obtained.The results show that the optimized heat treatment process parameters are annealing temperature of 768℃,holding time of 35 min and furnace cooling method.And the accuracy of the prediction result is verified by experiments.
作者
王凯
王荣吉
周童
彭松
Wang Kai;Wang Rongji;Zhou Tong;Peng Song(School of Mechanical and Electrical Engineering,Central South University of Forestry and Technology,Changsha Hunan 410004,China)
出处
《金属热处理》
CAS
CSCD
北大核心
2022年第9期31-35,共5页
Heat Treatment of Metals
基金
湖南省教育厅科学研究重点项目(14A157)。
关键词
TWIP钢
BP神经网络
遗传算法
热处理工艺
参数优化
TWIP steel
BP neural network
genetic algorithm
heat treatment process
parameters optimization
