基于BP神经网络算法的耐磨钢热处理工艺优化
发布时间：2018-10-10 12:48
【摘要】：以耐磨钢牌号、淬火温度、淬火冷却方式、回火温度和回火冷却方式作为输入层参数,以硬度作为输出层参数,采用BP神经网络算法构建了耐磨钢热处理工艺优化的BP神经网络模型,并进行了模型的预测和应用验证。结果表明,该模型的输出参数平均相对预测误差为2.2%,具有较好的预测能力和较高的预测精度。与生产线现用工艺相比,采用BP神经网络模型优化工艺热处理后的NM360、NM400、NM500耐磨钢的磨损体积分别减小26%、26%、28%。
[Abstract]:Taking wear-resistant steel grade, quenching temperature, quenching cooling mode, tempering temperature and tempering cooling method as input layer parameters and hardness as output layer parameters, The BP neural network model for heat treatment process optimization of wear-resistant steel was constructed by using BP neural network algorithm, and the prediction and application of the model were carried out. The results show that the average relative prediction error of the output parameters of the model is 2.2, which has better prediction ability and higher prediction accuracy. Compared with the current production line, the wear volume of NM360,NM400,NM500 wear-resistant steel after heat treatment was optimized by using BP neural network model. The wear volume of NM360,NM400,NM500 wear-resistant steel decreased by 26% and 28% respectively.
【作者单位】： 承德石油高等专科学校;
【基金】：国家高技术研究发展(863)计划项目(2004AA412030)
【分类号】：TG161
本文编号：2261824
[Abstract]:Taking wear-resistant steel grade, quenching temperature, quenching cooling mode, tempering temperature and tempering cooling method as input layer parameters and hardness as output layer parameters, The BP neural network model for heat treatment process optimization of wear-resistant steel was constructed by using BP neural network algorithm, and the prediction and application of the model were carried out. The results show that the average relative prediction error of the output parameters of the model is 2.2, which has better prediction ability and higher prediction accuracy. Compared with the current production line, the wear volume of NM360,NM400,NM500 wear-resistant steel after heat treatment was optimized by using BP neural network model. The wear volume of NM360,NM400,NM500 wear-resistant steel decreased by 26% and 28% respectively.
【作者单位】： 承德石油高等专科学校;
【基金】：国家高技术研究发展(863)计划项目(2004AA412030)
【分类号】：TG161
【相似文献】
相关期刊论文 前2条
1 王以恒;;基于BP神经网络算法的北京市地铁站应急疏散能力仿真评估模型[J];中国安全生产科学技术;2012年01期
2 ;[J];;年期
相关会议论文 前2条
1 彭本红;吕永成;黎雪霖;刘春霞;;基于改进的BP神经网络算法的区域生态环境可持续发展协调度评价研究[A];管理科学与系统科学研究新进展——第7届全国青年管理科学与系统科学学术会议论文集[C];2003年
2 朱小梅;杨先凤;吕南;;基于主成分分析的BP神经网络算法在软件风险识别中的应用研究[A];“中国视角的风险分析和危机反应”——中国灾害防御协会风险分析专业委员会第四届年会论文集[C];2010年
相关硕士学位论文 前1条
1 蔡金亮;BP神经网络算法在球磨机控制系统中的应用研究[D];华北电力大学（河北）;2010年
，本文编号：2261824
本文链接：https://www.wllwen.com/kejilunwen/jiagonggongyi/2261824.html
