正在加载图片...
DOI:10.13374/.issn1001-053x.2011.05.013 第33卷第5期 北京科技大学学报 Vol.33 No.5 2011年5月 Journal of University of Science and Technology Beijing May 2011 基于ANFIS方法的连续定向凝固BFel0-1-一1合金的 压缩流变应力模型 甘春雷刘雪峰黄海友谢建新区 北京科技大学材料先进制备技术教育部重点试验室,北京100083 ☒通信作者,E-mail:jxie@mater..usth.edu.cm 摘要以连续定向凝固柱状晶组织BF10-1-1合金在应变速率为0.01~10s·和变形温度为25-500℃条件下的压缩试 验所得实测数据为基础,采用自适应神经网络模糊推理系统(ANFIS)方法,建立了连续柱状晶组织BF10一1-1合金压缩变形 真应力与变形温度、应变速率和真应变关系的预测模型.结果表明:ANIS模型预测的流变应力值与试验值之间的平均误差 为0.75%,均方根误差为2.13,相关系数为0.9996,很好地反映了实际变形过程的特征,而在相同情况下采用传统回归模型 预测的平均误差为6.28%,表明ANFIS模型具有优良的预测精度 关键词铜镍合金;压缩变形:柱状晶粒:流变应力:数学模型:自适应神经网络模糊推理系统 分类号TG115.5:TG146.1 Compressive flow stress model of BFel0-14 alloy fabricated by continuous unidi- rectional solidification process using ANFIS GAN Chun-ei,LIU Xue-feng,HUANG Hai-you,XIE Jian-xin Key Laboratory of the Ministry of Education of China for Advanced Materials Processing,University of Science and Technology Beijing,Beijing 100083, China Corresponding author,E-mail:jxxie@mater.ustb.edu.cn ABSTRACT Based on the compression experimental data of BFel0--alloy with continuous unidirectionally solidified columnar grains,a prediction model for the relation of true stress to temperature,strain rate and true strain was developed using an adaptive net- work based fuzzy inference system (ANFIS).The temperature at which the alloy was compressed was from 25 to 500C with the strain rate ranging from 0.01 to 10s.Simulation results show that the mean percentage error,root mean square error and correlation coeffi- cient between the ANFIS model and measured data of flow stress are 0.75%,2.13 and 0.9996,respectively,indicating that the AN- FIS model can well reflect the real feature of the alloy during practical deforming process.In comparison with the regression model, whose mean percentage error is 6.28%under the same condition,ANFIS parades more accurate prediction performance for flow stress. KEY WORDS copper-nickel alloys:compression deformation:columnar grains;flow stress:mathematical models;adaptive network based fuzzy inference system 铜镍合金(白铜)具有优异的耐海水腐蚀和抗 凝固组织致密,塑性指标大幅度提高习.研究连续 海生物附着性能,在海滨电站、石油化工、船舶和海 柱状晶组织铜镍合金在不同温度下的流变应力变化 水淡化等领域应用较为广泛.采用传统的铸造方法 规律,建立预测模型(本构关系),对于进行材料成 制备铜镍合金容易产生缩孔、缩松和偏析等缺陷,影 形过程数值模拟和制定合理的成形加工工艺具有重 响材料的力学性能和耐腐蚀性能。研究发现,采用 要意义. 连续定向凝固方法可以制备性能优良的铜镍合金, 传统的流变应力预测模型的建立大多采用以试 收稿日期:2010-06-13 基金项目:国家自然科学基金重点资助项目(No.50674008)第 33 卷 第 5 期 2011 年 5 月 北京科技大学学报 Journal of University of Science and Technology Beijing Vol. 33 No. 5 May 2011 基于 ANFIS 方法的连续定向凝固 BFe10--1--1 合金的 压缩流变应力模型 甘春雷 刘雪峰 黄海友 谢建新 北京科技大学材料先进制备技术教育部重点试验室,北京 100083 通信作者,E-mail: jxxie@ mater. ustb. edu. cn 摘 要 以连续定向凝固柱状晶组织 BFe10--1--1 合金在应变速率为 0. 01 ~ 10 s - 1 和变形温度为 25 ~ 500 ℃ 条件下的压缩试 验所得实测数据为基础,采用自适应神经网络模糊推理系统( ANFIS) 方法,建立了连续柱状晶组织 BFe10--1--1 合金压缩变形 真应力与变形温度、应变速率和真应变关系的预测模型. 结果表明: ANFIS 模型预测的流变应力值与试验值之间的平均误差 为 0. 75% ,均方根误差为 2. 13,相关系数为 0. 999 6,很好地反映了实际变形过程的特征,而在相同情况下采用传统回归模型 预测的平均误差为 6. 28% ,表明 ANFIS 模型具有优良的预测精度. 关键词 铜镍合金; 压缩变形; 柱状晶粒; 流变应力; 数学模型; 自适应神经网络模糊推理系统 分类号 TG115. 5; TG146. 1 Compressive flow stress model of BFe10-1-1 alloy fabricated by continuous unidi￾rectional solidification process using ANFIS GAN Chun-lei,LIU Xue-feng,HUANG Hai-you,XIE Jian-xin Key Laboratory of the Ministry of Education of China for Advanced Materials Processing,University of Science and Technology Beijing,Beijing 100083, China Corresponding author,E-mail: jxxie@ mater. ustb. edu. cn ABSTRACT Based on the compression experimental data of BFe10-1-1 alloy with continuous unidirectionally solidified columnar grains,a prediction model for the relation of true stress to temperature,strain rate and true strain was developed using an adaptive net￾work based fuzzy inference system ( ANFIS) . The temperature at which the alloy was compressed was from 25 to 500 ℃ with the strain rate ranging from 0. 01 to 10 s - 1 . Simulation results show that the mean percentage error,root mean square error and correlation coeffi￾cient between the ANFIS model and measured data of flow stress are 0. 75% ,2. 13 and 0. 999 6,respectively,indicating that the AN￾FIS model can well reflect the real feature of the alloy during practical deforming process. In comparison with the regression model, whose mean percentage error is 6. 28% under the same condition,ANFIS parades more accurate prediction performance for flow stress. KEY WORDS copper-nickel alloys; compression deformation; columnar grains; flow stress; mathematical models; adaptive network based fuzzy inference system 收稿日期: 2010--06--13 基金项目: 国家自然科学基金重点资助项目( No. 50674008) 铜镍合金( 白铜) 具有优异的耐海水腐蚀和抗 海生物附着性能,在海滨电站、石油化工、船舶和海 水淡化等领域应用较为广泛. 采用传统的铸造方法 制备铜镍合金容易产生缩孔、缩松和偏析等缺陷,影 响材料的力学性能和耐腐蚀性能. 研究发现,采用 连续定向凝固方法可以制备性能优良的铜镍合金, 凝固组织致密,塑性指标大幅度提高[1--2]. 研究连续 柱状晶组织铜镍合金在不同温度下的流变应力变化 规律,建立预测模型( 本构关系) ,对于进行材料成 形过程数值模拟和制定合理的成形加工工艺具有重 要意义. 传统的流变应力预测模型的建立大多采用以试 DOI:10.13374/j.issn1001-053x.2011.05.013
向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有