Quality prediction of aluminum alloy resistance spot welding based on correlation analysis and SSA-BP neural network
-
摘要:
基于电阻点焊过程中工艺信号特征,在不同间距、不同间隙和不同间距与间隙3种条件下,引入相关性分析方法分析工艺信号与熔核直径之间的相关性,并建立基于麻雀搜索算法-BP神经网络(sparrow search algorithm- back propagation neural network, SSA-BP)的电阻点焊质量预测模型,将功率、焊接电流、焊接电压和动态电阻作为预测模型输入特征. 结果表明,经麻雀搜索算法优化后的BP神经网络在测试集上的决定系数R2、均方误差(mean-square error, MSE)、均方根误差(root mean square error, RMSE)和平均绝对误差(mean absolute error, MAE)分别为0.95,1.55,1.24和0.90,均优于BP模型. 获得了功率、焊接电流、焊接电压和动态电阻与熔核直径的映射关系,可为焊接的工艺参数设计提供依据.
Abstract:Based on the characteristics of the process signals in the resistance spot welding process, three working conditions of different spacing, different gaps and different spacing and gaps are analyzed, and correlation analysis is introduced to extract the correlation between the process signals and the diameter of nugget. A resistance spot welding quality prediction model based on Sparrow Search Algorithm-Back Propagation Neural Network (SSA-BP) was established, and power, welding current, welding voltage and dynamic resistance are used as input features of the prediction model. The results show that the BP neural network optimized by the sparrow search algorithm outperforms the BP model on the test set with R2, MSE, RMSE and MAE of 0.95, 1.55, 1.24 and 0.90, respectively. It is also determined that there exists a mapping relationship between power, welding current, welding voltage and dynamic resistance and the diameter of the nugget, which provides a basis for the design of process parameters for welding.
-
-
表 1 2219/5A06的力学性能
Table 1 Mechanical properties of 2219/5A06
材料 抗拉强度
Rm/MPa屈服强度
Rp0.2/MPa断后伸长率
A(%)弹性模量
E/GPa2219 455 350 10 73 5A06 315 255 20 71 表 2 2219/5A06的化学成分(质量分数,%)
Table 2 Chemical composition of 2219/5A06
材料 Al Si Fe Cu Mn 2219 92.8 0.06 0.17 6.30 0.31 5A06 93.31 0.06 0.13 0.03 — 材料 V Zr Zn Ti Cr 2219 0.1 0.15 0.02 0.07 — 5A06 — — 0.02 0.05 — 表 3 固定间距和不同间隙条件下焊点数量
Table 3 Fixed spacing and corresponding welding spots number under different gaps
装配条件 间隙G/mm 间距S/mm 焊点数量n(个) 1 0.1 50 15 2 0.2 50 15 3 0.3 50 15 4 0.4 50 15 5 0.5 50 15 6 0.6 50 15 7 0.8 50 15 8 1.0 50 15 9 1.5 50 15 表 4 固定间隙和不同间距条件下焊点数量
Table 4 Fixed gap and corresponding welding spots number under different spacing
装配条件 间隙G/mm 间距S/mm 焊点数量n(个) 10 0 10 15 11 0 15 15 12 0 20 15 13 0 25 15 14 0 30 15 15 0 35 15 16 0 40 15 17 0 45 15 18 0 50 15 表 5 不同间距和间隙条件下焊点数量
Table 5 The number of welding spots corresponding to different spacing and gaps
装配条件 间隙
G/mm间距
S/mm焊点数量
n(个)19 1 15 15 20 1 20 15 21 1 25 15 22 1 30 15 23 1.5 15 15 24 1.5 20 15 25 1.5 25 15 26 1.5 30 15 27 2 15 15 28 2 20 15 29 2 25 15 30 2 30 15 表 6 工艺信号与熔核直径之间的相关系数
Table 6 Correlation coefficient between process signal and nugget diameter
功率
P/kW焊接电流
I/kA焊接电压
U/mV动态电阻
R/μΩ0.62 0.95 0.65 0.54 表 7 预测模型精度指标
Table 7 Prediction model accuracy indicators
预测模型 评价指标 决定系数R2 均方根误差 RMSE 均方误差MSE 平均绝对误差MAE 改进SSA-BP 0.95 1.24 1.55 0.90 BP 0.79 2.57 1.88 1.41 SSA-BP 0.85 1.58 2.52 1.27 -
[1] 谢泽豪, 李建宇, 陈树海, 等. 钢/铝异种金属点焊研究进展综述[J]. 航空学报, 2022, 43(2): 43 − 57. Xie Zehao, Li Jianyu, Chen Shuhai, et al. Research progress in steel/Al-alloys dissimilar metals spot welding[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(2): 43 − 57.
[2] 雷达, 王海林, 周彪, 等. 铝合金-低碳钢异种金属电阻点焊工艺研究[J]. 材料导报, 2020, 34(S2): 465 − 468. Lei Da, Wang Hailin, Zhou Biao, et al. Research on resistance spot welding technology of aluminium alloy-low carbon steel dissimilar metal[J]. Materials Reports, 2020, 34(S2): 465 − 468.
[3] 夏裕俊, 李永兵, 楼铭, 等. 电阻点焊质量监控技术研究进展与分析[J]. 中国机械工程, 2020, 31(1): 100−125. Xia Yujun, Li Yongbin, Lou Ming, et al. Recent advances and analysis of quality monitoring and control technologies for RSW[J]. China Mechanical Engineering, 2020, 31(1): 100 − 125.
[4] 徐靖, 万晓东, 王元勋, 等. 基于电极电压信号的微电阻点焊质量监测分析[J]. 焊接学报, 2017, 38(9): 51 − 54. Xu Jing, Wan Xiaodong, Wang Yuanxun, et al. Electrode voltage based quality monitoring in small scale resistance spot welding[J]. Transactions of the China Welding Institution, 2017, 38(9): 51 − 54.
[5] 宇慧平, 元月, 韩长录, 等. 不同工艺下超强钢点焊残余应力的试验分析[J]. 焊接学报, 2016, 37(9): 35 − 38. Yu Huiping, Yuan Yue, Han Changlu, et al. Analysis of test about residual stress of super steel spot welding under different process[J]. Transactions of the China Welding Institution, 2016, 37(9): 35 − 38.
[6] Zhao D W, Wang Y X, Liang D Z, et al. Performances of regression model and artificial neural network in monitoring welding quality based on power signal[J]. Journal of Materials Research and Technology, 2019, 9(2): 1231 − 1240.
[7] Wan X, Wang Y, Zhao D. Quality evaluation in small-scale resistance spot welding by electrode voltage recognition[J]. Science and Technology of Welding and Joining, 2016, 21(5): 358 − 65. doi: 10.1080/13621718.2015.1115161
[8] 赵大伟, 王新阳, 王元勋. 钛合金微电阻点焊电极间电压质量检测技术[J]. 焊接学报, 2014, 35(1): 33 − 36. Zhao Dawei, Wang Xinyang, Wang Yuanxun. Electrode voltage quality detection technology of titanium alloy micro resistance spot welding[J]. Transactions of the China Welding Institution, 2014, 35(1): 33 − 36.
[9] Pashazadeh H, Gheisari Y, Hamedi M. Statistical modeling and optimization of resistance spot welding process parameters using neural networks and multi-objective genetic algorithm[J]. Journal of Intelligent Manufacturing, 2016, 27(3): 549 − 559. doi: 10.1007/s10845-014-0891-x
[10] 欧阳城添, 朱东林, 邱亚娴. 融合聚类算法的改进麻雀搜索算法[J]. 计算机仿真, 2022, 39(12): 392 − 397. Ouyang Chengtian, Zhu Donglin, Qiu Yaxian. Improved sparrow search algorithm based on clustering algorithm[J]. Computer Simulation, 2022, 39(12): 392 − 397.
[11] 苏莹莹, 王升旭. 自适应混合策略麻雀搜索算法[J]. 计算机工程与应用, 2023, 59(9): 75 − 85. Su Yingying, Wang Shengxu. Adaptive hybrid strategy sparrow search algorithm[J]. Computer Engineering and Applications, 2023, 59(9): 75 − 85.
[12] Xue J, Shen B. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems Science& Control Engineering, 2020, 8(1): 22 − 34.
[13] Gharehchopogh F S, Namazi M, Ebrahimi L, et al. Advances in sparrow search algorithm: a comprehensive survey[J]. Archives of Computational Methods in Engineering, 2023, 30(1): 427 − 455. doi: 10.1007/s11831-022-09804-w
[14] 闫鹏程, 尚松行, 张超银, 等. 改进BP神经网络算法对煤矿水源的分类研究[J]. 光谱学与光谱分析, 2021, 41(7): 2288 − 2293. Yan Pengcheng, Shang Songxing, Zhang Chaoyin, et al. Study on classification of coal mine water source by improved BP neural network algorithm[J]. Spectroscopy and Spectral Analysis, 2021, 41(7): 2288 − 2293.
[15] Tang X Y, Feng D F, Li K Q, et al. An improved BPNN prediction method based on multi-strategy sparrow search algorithm[J]. Computers, Materials and Continua, 2023, 74(2): 2789 − 2802. doi: 10.32604/cmc.2023.031304
[16] 彭业飞, 杨露菁, 黄璜. 基于Logistic和Tent双重映射的混沌粒子群算法[J]. 数字技术与应用, 2015(12): 136 − 137. Peng Yefei, Yang Lujing, Huang Huang. Logistic and tent chaotic particle swarm optimization[J]. Digital Technology & Applicationn, 2015(12): 136 − 137.