Wang, M., Yang, H., Li, Y., Zhou, J., Huang, Q., Yao, X.
42062424100;55576731200;55892154700;55493538900;55891834200;55747884900;
Elimination of voids in crankshaft through a hybrid of back propagation neural network and genetic algorithm
(2013) Huagong Xuebao/CIESC Journal, 64 (10), pp. 3673-3678.
https://www.scopus.com/inward/record.uri?eid=2-s2.0-84885909848&doi=10.3969%2fj.issn.0438-1157.2013.10.026&partnerID=40&md5=ef33d104d030c81c73ac973add51e348

DOI: 10.3969/j.issn.0438-1157.2013.10.026
归属机构: School of Material Science and Engineering, Chongqing University, Chongqing 400030, China;
Gree Electric Appliances, Ltd. of Chongqing, Chongqing 400039, China
摘要: A method of combining back propagation neural network (BP neural network) and genetic algorithm was proposed to optimize the process parameters and eliminate the voids in crankshaft. Mold temperature, melt temperature, packing pressure and gate size were taken as design variables and sink marks were taken as optimization goal. Computer aided engineering (CAE) simulation was performed based on the Taguchi method. A BP neural network model was developed to obtain the mathematical relationship between optimization goal and design variables, and genetic algorithm was used to optimize the process parameters. The optimal process parameters were mold temperature 80°C, melt temperature 210°C, packing pressure 110 MPa, gate size 1mm. Finally, the voids in the crankshaft could be eliminated by using the optimized process parameters in actual factory production. © All Rights Reserved.
作者关键字: Back propagation neural network;  Crankshaft;  Genetic algorithm;  Taguchi method;  Voids
索引关键字: Back propagation neural networks; BP neural network model; BP neural networks; Computer aided engineering simulations; Mathematical relationship; Optimization goals; Process parameters; Voids, Computer aided engineering; Crankshafts; Molds; Neural networks; Optimization; Taguchi methods, Genetic algorithms
通讯地址: Yang, H.; School of Material Science and Engineering, Chongqing University, Chongqing 400030, China; 电子邮件: yang_hai200888@126.com
ISSN: 04381157
CODEN: HUKHA
原始文献语言: Chinese
来源出版物名称缩写: Huagong Xuebao
文献类型: Article
来源出版物: Scopus