[1]何 帅,王立君,梁恩宝,等.预测X65钢堆焊质量的PSO+BP算法[J].焊管,2015,38(2):5-10.[doi:１００１－３９３８（２０１5）02-0005-06]
　HE Shuai,WANG Lijun,LIANG Enbao.PSO + BP Algorithm Prediction of X65 Steel Surfacing Quality[J].,2015,38(2):5-10.[doi:１００１－３９３８（２０１5）02-0005-06]
点击复制

预测X65钢堆焊质量的PSO+BP算法
分享到：

《焊管》[ISSN:1001-3938/CN:61-1160/TE]

卷:
38
期数:
2015年第2期
页码:
5-10
栏目:
试验与研究
出版日期:
2015-02-28

文章信息/Info

Title:
PSO + BP Algorithm Prediction of X65 Steel Surfacing Quality
文章编号:
１００１－３９３８（２０１5）02-0005-06
作者:
何 帅1; 2; 王立君1; 2; 梁恩宝1
（1.天津大学 材料科学与工程学院，天津 300072；
2. 天津市现代连接技术重点实验室，天津 300072）
Author(s):
HE Shuai1; 2;  WANG Lijun1; 2;  LIANG Enbao1; 2
(1.School of Material Science and Engineering, Tianjin University, Tianjin 300072, China;
(2.Key Laboratory of Advanced Joining Technology, Tianjin 300072, China)
关键词:
焊接; 堆焊; 质量预测; 粒子群算法; 神经网络
Keywords:
welding;  surfacing;  quality prediction;  particle swarm optimization(PSO);  neural network
分类号:
TG455
DOI:
１００１－３９３８（２０１5）02-0005-06
文献标志码:
A
摘要:
现行的焊接工艺预测算法难以满足算法设计的适用性、可靠性以及高效性的原则。根据工程需要，建立了以电弧长度、焊接电流、焊接速度、送丝速度和保护气流量为输入，以堆焊后的熔宽、熔深和稀释率为输出的5-8-3结构的误差反向传播(BP)网络模型，利用粒子群算法(PSO)优化BP网络得到最优权值和阈值来预测X65钢板堆焊Inconel625镍基合金的焊后质量。结果表明，PSO+BP算法相比单一BP算法具有较高的准确性，比遗传优化BP网络(GA+BP)算法高效。与GA+BP算法相比，稀释率的平均误差分别为0.30和1.05，计算时间分别6 726 s和11 034 s，将PSO优化后的最优权值与Chebyshev直接法确定的权值对比，得出两个模型的权值基本吻合，说明PSO+BP算法预测堆焊质量过程中没有陷入局部最优解，具有准确、高效和可靠的优点，适用于堆焊质量的预测。
Abstract:
The current welding process prediction algorithm cannot satisfy the applicability, reliability and efficiency of algorithm design principles. According to the requirements of the project, set up the error Back Propagation(BP) network model of the 5-8-3 structure with arc length, welding current, welding speed and wire feed speed and protection gas flow rate as input, weld width and weld height and the dilution rate after surfacing as output. Particle Swarm Optimization(PSO) was used to optimize the BP network to get the optimal weights and threshold to predict the quality of Inconel625 nickel base alloy surfacing X65 steel after welding. The results showed that the PSO+BP algorithm has higher accuracy compared with the single BP algorithm, and is more efficient than Genetic optimizing BP network(GA+BP) algorithm. Compared with the GA+BP algorithm, the dilution rate of average error and computing time were 0.30, 1.05 and the 6 726 s and 11 034 s. To contrast the optimal weights are optimized by PSO and Chebyshev direct method, it is concluded that the PSO+BP algorithm has not trapped in local optimal solution, and has the advantages of accurate, efficient and reliable, is suitable for welding quality prediction.

参考文献/References:

［1］杨洁. 30CrMo钢表面堆焊Inconel625镍基合金性能的研究[D]. 兰州：兰州理工大学，2011.
［2］李玉龙，禹业晓. TIG堆焊技术研究进展[J]. 电焊机，2012，42(12)：70-76.
［3］徐玉强，徐晓明，许可望，等. TIP TIG焊接技术在双金属复合管海底管道铺设中的应用[J]. 焊接与切割，2013(06)：18-19.
［4］刘衍民，牛奔. 新型粒子群算法理论与实践[M]. 北京: 科学出版社，2013.
［5］彭金宁，陈丙森，朱平. 焊接工艺参数的神经网络智能设计[J]. 焊接学报，1998，19(01)：19-24.
［6］叶建雄，张晨曙. 焊接工艺参数中的BP神经网络与遗传算法结合[J]. 上海交通大学学报，2008，42(11)：57-60.
［7］刘波. 粒子群优化算法及其工程应用[M]. 北京：电子工业出版社，2010.
［8］史峰，王小川. MATLAB神经网络43个案例分析[M].北京：北京航空航天大学出版社，2013.
［9］YUNONG Z, TONGKE Z, WEI L. Growing algorithm of laguerre orthogonal basis neural network with weights directly determined[C]//Proceedings of International Conference on Intelligent Computing. Shanghai: [s.n.], 2008: 60-67.
[10］张雨浓，肖秀春，陈扬文, 等. Hermite前向神经网络隐节点数目自动确定[J]. 浙江大学学报（工学版），2010，44(02): 23-26.
［11］ ZHANG Y, LI W, CHEN K. A weight-directly-determined simple neural network for nonlinear system identification[C]//Proceedings of IEEE International Conference on Fuzzy Systems. Washington D.C: [s.n.] , 2008: 455-460.
［12］ ZHANG Y, RUAN G. Bernoulli neural network with weights directly determined and with the number of hidden-layer neurons automatically determined[C]// Proceedings of International Symposium on Neural Networks. Wuhan: [s.n.]，2009: 36-45.
［13］ BANOVIC S W, DUPONT J N, MARDER A R. Dilution control in gas-tungsten-arc welds involving superaustenitic stainless steels and nickel-based alloys [J]. Metallurgical and Materials Transactions, 2001, 32(12): 33-36.
［14］ KENNEDY J, EBERHART C. Particle swarm Opitimization[C]//In Proceedings of IEEE International Conference on Neural Networks. Washington:[s.n.], 1995: 1942-1948.
［15］ WEI X, PAN H, MA Q. Application of particle swarm optimization based neural network to fault diagnosis [J]. Journal of Vibration, Measurement & Diagnosis, 2006, 26(02): 133-137.
［16］ BO L, HONGXIA P. A hybrid PSO-DV based intelligent method for fault diagnosis of gear-box[C]//2009 IEEE International Symposium on Computational Intelligence in Robotics and Automation(CIRA2009).Deajeon:[s.n.], 2009.
［17］ SHI Y, EBERHART R C. Empirical study of particle swarm optimization[C]// 1999 Congress on Evolutionary Computing. Washington D.C: [s.n.], 1999: 1945-1950.
［18］ ANDREWS P S. An investigation into mutation operators for particle swarm optimization [C]// In IEEE Congress on Evolution Computation. Vancouver:[s.n.], 2006: 1029-1036.
［19］ 邹阿金，张雨浓. 基函数神经网络及应用[M]. 广州：中山大学出版社，2009.
［20］张雨浓，李巍，蔡炳煌，等. 切比雪夫正交基神经网络的权值直接确定法[J]. 计算机仿真, 2009, 26(01)：157-161.
［21］ MATHEWS J H，FINK K D. Numerical Methods Using MATLAB[M]. Beijing: Person Education Inc., 2004.
［22］张雨浓，杨逸文，李巍. 神经网络权值直接确定法[M]. 广州：中山大学出版社，2010.

相似文献/References:

[1]赵春友.中直径直缝双面埋弧焊钢管[J].焊管,2007,30(5):59.[doi:1001－3938（2007）05－0059－02]
　ZHAO Chun-you.Longitudinal Double Submerged Arc Welded Pipe with Medium Diameter[J].,2007,30(2):59.[doi:1001－3938（2007）05－0059－02]
[2]郭瑞杰,付桂英,杨天冰,等.长输油气管道激光焊接技术[J].焊管,2007,30(6):78.[doi:1001－3938（2007）06－0078－03]
　GUO Rui-jie,FU Gui-ying.Laser Welding Technology of Long Distance Pipeline[J].,2007,30(2):78.[doi:1001－3938（2007）06－0078－03]
[3]毕宗岳,张锦刚.连续管焊丝的试验研究[J].焊管,2008,31(1):26.[doi:1001－3938（2008）01－0026－03]
　BI Zong-yue,ZHANG Jin-gang.Testing & Research of Welding Wire Used for Coil Tubing[J].,2008,31(2):26.[doi:1001－3938（2008）01－0026－03]
[4]韩志诚,王少刚,徐凤林,等.双相不锈钢的焊接研究进展[J].焊管,2008,31(3):5.[doi:1001－3938（2008）03－0005－04]
　HAN Zhi-cheng,WANG Shao-gang,XU Feng-lin,et al.Progress of Research on Welding of Duplex Stainless Steel[J].,2008,31(2):5.[doi:1001－3938（2008）03－0005－04]
[5]张盟军,刘换军,张 璐,等.林肯焊机并联技术的应用[J].焊管,2008,31(3):56.[doi:1001－3938（2008）03－0056－02]
　ZHANG Meng-jun,LIU Huan-jun,ZHANG Lu,et al.Parallel Connection Technology Application of Lincoln Welding Machine[J].,2008,31(2):56.[doi:1001－3938（2008）03－0056－02]
[6]周友龙,方永龙,袁卫东,等.T91/TP347H管接头焊接工艺试验研究[J].焊管,2008,31(4):17.[doi:1001－3938（2008）04－0017－04]
　ZHOU You-long,FANG Yong-long,YUAN Wei-dong,et al.Study on Welding Procedure Test of T91/TP347H Pipe Joint[J].,2008,31(2):17.[doi:1001－3938（2008）04－0017－04]
[7]冯 斌,刘 宇,刘方明.X70级管线钢焊接热裂纹模拟研究[J].焊管,2009,32(1):15.[doi:1001－3938(2009)01－0015-06]
　FENG Bin,LIU Yu,LIU Fang-min.Welding Thermal Crack Simulation Study on X70 Grade Pipeline Steel[J].,2009,32(2):15.[doi:1001－3938(2009)01－0015-06]
[8]徐 刚,王 旭,李树军,等.林肯交流焊机斯考特连接的研究[J].焊管,2009,32(3):52.[doi:1001－3938（2009）03－0052－02]
　XU Gang,WANG Xu,LI Shu-jun,et al.The Research of Lincoln AC Welding Machine Scott Connection[J].,2009,32(2):52.[doi:1001－3938（2009）03－0052－02]
[9]张祥,曾涛.基于ANSYS压力容器筒体与平板封头焊缝残余应力有限元分析[J].焊管,2009,32(5):22.[doi:1001－3938（2009）05－0022－03]
　ZHANG Xiang,ZENG Tao.The Finite Element Analysis on the Weld Residual Stress of the Pressure Vessel Cylinder and Plate Dome Based on ANSYS[J].,2009,32(2):22.[doi:1001－3938（2009）05－0022－03]
[10]尹长华,邹欣,李广民.“东西伯利亚—太平洋”管道工程项目中焊接相关认证介绍[J].焊管,2009,32(6):59.[doi:1001－3938（2009）06－0059－05]
　YIN Chang-hua,ZOU Xin,LI Guang-min.Regarding qualification and certificate Correlated with Welding[J].,2009,32(2):59.[doi:1001－3938（2009）06－0059－05]
[11]左丽娜,杜先鹏,马玉强,等.大型阀门耐腐蚀层的堆焊与加工工艺研究[J].焊管,2014,37(3):34.[doi:１００１－３９３８（２０１4）03-0034-05]
　ZUO Lina,DU Xianpeng,MA Yuqiang,et al.Research on Corrosion Resistance Layers Surfacing and Cutting Processes of Large-scale Valve[J].,2014,37(2):34.[doi:１００１－３９３８（２０１4）03-0034-05]
[12]杨 玭,曹晶晶,袁鹏斌.316L/X65双金属复合管管端根焊工艺研究[J].焊管,2015,38(12):40.[doi:１００１－３９３８（２０１5）12-0040-06]
　YANG Pin,CAO Jingjing,YUAN Pengbin.Research on Root Welding Process of 316L/X65 Double-metal Composite Pipe End[J].,2015,38(2):40.[doi:１００１－３９３８（２０１5）12-0040-06]
[13]吴建英.细长喷水管道的堆焊工艺及变形控制[J].焊管,2017,40(9):39.[doi:10.19291/j.cnki.1001-3938.2017.09.009]
　WU Jianying.Surfacing Welding Process and Deformation Control of Elongated Spary Pipe[J].,2017,40(2):39.[doi:10.19291/j.cnki.1001-3938.2017.09.009]

备注/Memo

备注/Memo:
收稿日期：2014-10-22
基金项目：天津市科技支撑计划重点项目（11ZCGYSF00100）。
作者简介：何帅（1989—），男，硕士研究生，研究方向为焊接过程的智能控制及数值模拟。
更新日期/Last Update: