基于神经网络的焊机参数预测方法
Prediction method of welding machine parameters based on neural network
-
摘要: 针对脉冲MIG焊参数众多,不易调节的特点,提出了一种基于神经网络的焊机参数预测方法. 该方法采用LM(levenberg-marquarlt)算法建立了焊机参数的BP(back propagation)神经网络模型,充分利用已知的理想数据对网络进行训练,实现了焊接过程中任一给定焊接电流状态下焊机输出参数的预测;利用焊接参数的预测值分别对单、双脉冲MIG焊进行了试焊. 结果表明,基于神经网络的焊机参数预测方法精度较高,焊接过程稳定,焊缝成形美观,能够实现良好的一元化调节.Abstract: In view of the fact that pulse MIG welding has many parameters and is difficult to adjust, a welding parameter prediction method based on neural network is proposed. This method, having established BP neural network model of welding parameters by adopting LM(levenberg-marquarlt) algorithm, and making full use of the known data to train the network, have realized the prediction of the output parameters in any given welding current state, and then conduct test weld on single and double pulse MIG welding respectively by using the predicted values of welding parameters. The results show that the prediction method of welding parameters based on neural network is of high accuracy, that the welding process is stable, and that the seams can be well-formed, thus achieving a good unified adjustment.
-
Keywords:
- neural network /
- prediction method /
- unified adjustment /
- pluse welding
-
-
[1] Sen M, Mukherjee M, Pal T K. Evaluation of correlations between DP-GMAW process parameters and bead geometry[J]. Welding Journal, 2015, 94(8): 265s-279s.[2] Jie Y I, CAO S, LI L,et al. Effect of welding current on morphology and microstructure of Al alloy T-joint in double-pulsed MIG welding[J]. Transactions of Nonferrous Metals Society of China, 2015, 25(10): 3204-3211.[3] 林 放, 黄文超, 陈小峰, 等. 基于局部牛顿插值的数字化焊机参数自调节算法[J]. 焊接学报, 2011, 32(3): 33-36.Lin Fang, Huang Wenchao, Chen Xiaofeng,et al. Digitial welder parameters self-regulating algorithm based on partial Newton interPolation[J]. Transactions of the China Welding Insititution, 2011, 32(3): 33-36.[4] 薛家祥, 姜乘风, 张晓莉, 等. 基于最小二乘法的脉冲 MIG 焊参数一元化调节[J]. 焊接学报, 2014, 35(8): 75-78.Xue Jiaxiang, Jiang Chengfeng, Zhang Xiaoli,et al. Research on unified adjustment of pulsed MIG welding parameters based on least squares method[J]. Transactions of the China Welding Insititution, 2014, 35(8): 75-78.[5] 黄文超, 熊丹枫, 薛家祥. 铝硅合金双脉冲MIG焊专家数据库[J]. 焊接技术, 2009, 38(11): 43-46.Huang Wenchao, Xiong Danfeng, Xue Jiaxiang. Study on expert database of double pulse MIG welding for Al-Si alloy[J]. Welding Technology, 2009, 38(11): 43-46.[6] 陈小峰, 林 放, 魏仲华, 等. 基于数学建模的铝合金双脉冲MIG焊专家数据库设计[J]. 焊接学报, 2011, 32(5): 37-40.Cheng Xiaofeng, Lin Fang, Wei Zhonghua,et al. Double-pulsed MIG expert database based on mathematical modeling[J]. Transactions of the China Welding Insititution, 2011, 32(5): 37-40.[7] 刘 倩. 基于人工神经网络的电池容量预测[J]. 武汉理工大学学报, 2006, 28(3): 28-31.Liu Qian. Estimation for SOC of MH/Ni battery based on artificial neural network[J]. Journal of Wuhan University of Technology, 2006, 28(3): 28-31. -
期刊类型引用(6)
1. 卓文波,谭国笔,陈秋任,侯泽宏,王显会,韩维建,黄理. 基于代理模型和NSGA-Ⅱ的超高强钢电阻点焊工艺参数多目标优化. 焊接学报. 2024(04): 20-25+130 . 本站查看
2. 梁涛,刘锋,王睿,高忠林. 基于PLC的多通信方式焊机测试系统. 热加工工艺. 2021(05): 146-150 . 百度学术
3. 黄无云,刘益剑,刘宗熙,杨继全,朱钊伟,谢非,史建军. 焊道形貌特征的BP神经网络建模与预测. 南京师范大学学报(工程技术版). 2021(01): 1-7 . 百度学术
4. 吕小青,王旭,徐连勇,荆洪阳,韩永典. 基于组合模型的MAG焊工艺参数多目标优化. 焊接学报. 2020(02): 6-11+97 . 本站查看
5. 吕小青,王旭,徐连勇,荆洪阳. 基于径向基函数神经网络和NSGA-Ⅱ的气保焊工艺多目标优化. 天津大学学报(自然科学与工程技术版). 2020(10): 1013-1018 . 百度学术
6. 胡为,常新新,姬书得,李峰,宋崎,牛士玉. 基于RBF-GA的铝/镁异材FSLW工艺参数优化. 焊接学报. 2020(06): 54-59+84+100 . 本站查看
其他类型引用(7)
计量
- 文章访问数: 758
- HTML全文浏览量: 17
- PDF下载量: 2
- 被引次数: 13