基于数据模型与决策优化的高温合金铸件冒口设计  被引量:2

Feeding System Design Based on Data Model and Decision Optimization for Superalloy Castings

在线阅读下载全文

作  者:孙锦玉[1] 汪东红 刘淑梅[1] 疏达[2,3] Sun Jinyu;Wang Donghong;Liu Shumei;Shu Da(School of Materials Engineering,Shanghai University of Engineering Science;School of Materials Science and Engineering,Shanghai Jiaotong University;Shanghai Key Laboratory of Advanced High-temperature Materials and Precision Forming)

机构地区:[1]上海工程技术大学材料工程学院 [2]上海交通大学材料科学与工程学院 [3]上海市先进高温材料及其精密成形重点实验室

出  处:《特种铸造及有色合金》2019年第1期23-27,共5页Special Casting & Nonferrous Alloys

基  金:国家重点研发计划资助项目(2016YFB0701405);国家自然科学基金资助项目(51705314);上海市科委研发平台专项基金资助项目(16DZ2260602)

摘  要:以高温合金Inconel 939环套环铸件的冒口尺寸为优化对象,以铸件工艺出品率及缩松与铸件表面距离作为优化目标,采用均匀拉丁超立方法设计模拟方案,并根据模拟结果以不同的算法建立响应面模型,选取较优响应面模型,通过NSGA-Ⅱ算法对高温合金铸件冒口尺寸进行多目标优化,利用多准则决策MCDM(Multi Criteria Decision Maker)对多目标优化问题进行决策。比较数字化模型优化结果与模拟结果,发现铸件工艺出品率相差0.28%,内、外环冒口缩松位置与铸件表面距离分别为0.91、0.21mm。Taking the riser size of the superalloy Inconel 939 ring to ring castings as the optimized object,the casting yield and distance between the shrinkage and the casting surface were optimized.Firstly,the simulation scheme was designed by the uniform Latin Hypercube method.Then,the response surface model was established by different algorithms according to the simulated data.Finally,the optimal response surface model was selected to take multi-objective optimization of the riser size of the superalloy casting by the NSGA-Ⅱalgorithm.MCDM(multi criteria decision maker)was adopted to make decisions on multi-objective optimization problems.Through comparing the numerical model optimization results with the simulated ones,the casting yield difference is0.28%,and distance between the inner and outer ring riser shrinkage porosity and casting surface are 0.91 mm and 0.21 mm,respectively.

关 键 词:冒口优化 高温合金 响应面 NAGA-Ⅱ算法 MCDM 

分 类 号:TG249.5[机械工程—精密仪器及机械] TP311[金属学及工艺—铸造]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象