Energy control has been recognized as a core focus in welding research, and the development of a high-accuracy energy prediction model is considered essential for achieving precise energy output in ultrasonic welding. In response to the shortcomings of traditional energy prediction methods, such as difficulties in feature selection, high sensitivity to outliers, and limited prediction accuracy, a prediction model architecture integrating the convolutional neural network (CNN), long short-term memory network (LSTM), and attention mechanism was proposed. A comprehensive review of the physical modeling principles, model design details, and key technologies in ultrasonic energy prediction was presented, covering 1D-CNN feature extraction, bidirectional LSTM temporal modeling, and the design of an attention-enhanced layer for multi-mechanism fusion. The advantages of the CNN-LSTM-Attention model in terms of prediction performance, stability, and generalization ability were analyzed in detail. Experimental results show that this model exhibits outstanding performance in various operating conditions and complex environments, with root mean square error (RMSE) and mean absolute error (MAE) significantly superior to traditional deep learning models.
LIU Ning-zhuang, ZHAO Pei-yang, TIAN Hai-bo.
Energy Prediction Model for Ultrasonic Welding Based on Bayesian Optimization CNN-LSTM Attention[J].
Science Technology and Engineering, 2026, 26(25): 10944-10953 DOI:10.12404/j.issn.1671-1815.2506951
LiHuan, LiJun, LiZhe, et al. Plastic deformation and microstructure evolution in pure copper during ultrasonic welding[J]. Transactions of the China Welding Institution, 2025, 46(9): 93-101.
[7]
KirbyM, NaderiA, PalardyG. Predictive thermal modeling and characterization of ultrasonic consolidation process for thermoplastic composites[J]. Journal of Manufacturing Science and Engineering, 2023, 145(3): 13.
[8]
KimB J, HwangI, SonS G, et al. Review on ultrasonic welding quality monitoring technology[J]. Journal of Welding and Joining, 2024, 42(2): 215-222.
ZhangZhifen, ChenShanben, WenGuangrui. Research progress and prospect of welding intelligent monitoring technology[J]. Transactions of the China Welding Institution, 2024, 45(11): 10-20, 70.
[11]
FlorianW M, MirzC, WeilS, et al. Weld quality characterization by vibration analysis for ultrasonic metal welding processes[J]. Journal of Advanced Joining Processes, 2023, 8: 100149.
[12]
WuY, MengY, ShaoC. End-to-end online quality prediction for ultrasonic metal welding using sensor fusion and deep learning[J]. Journal of Manufacturing Processes, 2022. DOI: 10.1016/j.jmapro.2022.09.011.
[13]
HongS M, ChoJ W, BangH S. Quality classification of multi-layer copper foil stacks welds for ultrasonic welding using CNN and RNN[J]. Journal of Welding and Joining, 2024, 42(5): 514-520.
[14]
MohandasR, MonganP, HayesM. Ultrasonic weld quality inspection involving strength prediction and defect detection in data-constrained training environments[J]. Sensors, 2024, 24: 6553.
[15]
FlorianW M, MirzC, SchiebahnA, et al. Influence of quality features, disturbances, sensor data, and measurement time on quality prediction for ultrasonic metal welding[J]. Welding in the World, 2025. DOI: 10.1007/s40194-025-01959-x.
LiZidong, ZhangYong, ZhangZhonghao, et al. Prediction of frequency hopping sequences based on CNN-LSTM with attention mechanism[J/OL]. Radio Communication Technology, 1-1 5[2026-06-25]. https://link.cnki.net/urlid/13.1099.TN.20250924.1855.012.
LiShiwan, YuanMingdao, XuYunqian, et al. A CNN-LSTM model for predicting seepage flow of earth-rock dams based on attention mechanism[J]. Science Technology and Engineering, 2025, 25(21): 9102-9108.
[20]
杨庭飞. 塑料构件高效可靠超声波焊接成热机理与应用研究[D]. 扬州: 扬州大学, 2019.
[21]
YangTingfei. Research on the thermal mechanism and application of high-efficiency and reliable ultrasonic welding of plastic components[D]. Yangzhou: Yangzhou University, 2019.
WangJie, ZhangZhifen, BaiZijian, et al. Monitoring of welding forming quality based on hybrid driving of CNN-LSTM[J]. Transactions of the China Welding Institution, 2024, 45(11): 121-127.
WangKaisong, WeiYiming, TangWei, et al. Prediction method for pipe blockage faults in wet spray truck pumping system based on CNN-LSTM-Attention model[J]. Journal of Engineering Design, 2025, 32(6): 759-768.
ZhuAnfeng, ZhaoQiancheng, ZhouLing, et al. Wind turbine condition monitoring and health assessment based on CNN-LSTM-Attention[J]. Journal of Vibration, Measurement & Diagnosis, 2025, 45(2): 256-263, 409.
TangQingwei, XiangYue, DaiJiakun, et al. Prediction method for wind farm power generation migration based on CNN-LSTM[J]. Engineering Science and Technology, 2024, 56(2): 91-99.
YuWei, HuangHao, JinChenxing, et al. Ultra-short-term wind power prediction based on multi-head self-attention feature transformation and CNN-LSTM[J]. Southern Power System Technology, 2026, 20(5): 71-80, 102.
LeiKesong, Tusongjiang·Kari, Yilihamu·Yaermaimaiti, et al. Short-term photovoltaic power prediction based on WGAN-GP and CNN-LSTM-Attention[J]. Power System Protection and Control, 2023, 51(9): 108-118.
YangYong, KeTing, HuQizheng, et al. Fault diagnosis method for ZPW-2000A track circuit based on CNN-LSTM-Attention[J]. Journal of Railway Science and Engineering, 2025, 22(5): 2380-2392.
WangMin, TianDaping. Application of improved particle swarm optimization algorithm to optimize CNN LSTM Attention model in prediction of safety production accidents[J]. Journal of Safety and Environment, 2025, 25(5): 1829-1837.
JiXiaofeng, GuoYashi, ChenFang, et al. Tourism passenger flow prediction during golden week based on the combined model of CNN-LSTM-Attention: a case study of Dali Prefecture[J]. Arid Zone Resources and Environment, 2025, 39(3): 200-208.
YinGang, ZhuMiao, QuanPengcheng, et al. Research on prediction method of electrolysis temperature in aluminum electrolytic cells based on PID search optimization and CNN-LSTM-Attention[J]. Journal of Instrumentation and Measurement, 2025, 46(1): 324-337.
QiShuai, WangChao, JinZijun, et al. Combined model for intelligent prediction of slope stability and its application[J]. Science Technology and Engineering, 2025, 25(33): 14168-14176.
[46]
KuoC C, TsaiQ Z, LiD Y, et al. Optimization of ultrasonic welding process parameters to enhance weld strength of 3C power cases using a design of experiments approach[J]. Polymers, 2022, 14: 2388.