Shockwave signal reconstruction method based on a serial-parallel double branch network

SUN Chuanmeng1,2,CHEN Jiaxin1,2,YUAN Yue1,2,PEI Dongxing1,2,MA Tiehua1,2

Journal of Vibration and Shock ›› 2024, Vol. 43 ›› Issue (6) : 38-49.

PDF(3700 KB)
PDF(3700 KB)
Journal of Vibration and Shock ›› 2024, Vol. 43 ›› Issue (6) : 38-49.

Shockwave signal reconstruction method based on a serial-parallel double branch network

  • SUN Chuanmeng1,2,CHEN Jiaxin1,2,YUAN Yue1,2,PEI Dongxing1,2,MA Tiehua1,2
Author information +
History +

Abstract

Reconstructing the pressure distribution in the shock wave field by limited measurement point data and the complete shock wave pressure curve by missing data are of great value for weapon power and target damage assessment. To address the problem of reconstruction of the shockwave signal, it is proposed to establish the Res-GRU branch to capture the local timing dependence of the shockwave overpressure signal in a serial manner; to establish the Transformer branch to analyze the global potential features of the signal in a parallel manner; to establish the feature merging unit for higher-order feature integration to realize the complementary information of different stages layer by layer. Then, it is constructed a serial-parallel double branch network (denoted as G-TNet) based on GRU and Transformer model. The experimental study shows that G-TNet integrates the signal timing relationship, data variation pattern and other characteristic information; in the reconstruction experiment of pressure distribution of shockwave field based on limited measurement point data, the MSE between the reconstructed simulated and measured overpressure data and the original value were 5.0×10-6 and 1.2×10-3, the average peak error were 0.49% and 27.01%, and the average positive pressure time errors were 15.62% and 15.91%, and the average specific impulse errors were 17.66% and 19.33%, respectively; in the reconstruction experiments of the shockwave pressure curve based on the missing data, the MSE between the reconstructed simulated and measured signals and the original values were 5.0×10-6 and 5.0×10-4, and the MAE were 0.0010 and 0.0171, respectively; The G-TNet reconstruction results are better than the mainstream methods and fulfill the requirements of explosion shockwave pressure reconstruction index.

Key words

Dynamic test / Shock wave overpressure / Signal reconstruction / Deep learning / Feature merging

Cite this article

Download Citations
SUN Chuanmeng1,2,CHEN Jiaxin1,2,YUAN Yue1,2,PEI Dongxing1,2,MA Tiehua1,2. Shockwave signal reconstruction method based on a serial-parallel double branch network[J]. Journal of Vibration and Shock, 2024, 43(6): 38-49

References

[1] 祖静,马铁华,裴东兴,等.新概念动态测试[M].国防工业出版社,2016. Zu J, Ma TH, Pei DX, et al. New concept dynamic testing [M]. Defense Industry Press, 2016. [2] 徐浩,杜红棉,范锦彪,等.冲击波测试系统低频特性与补偿方法研究[J].爆炸与冲击,2019,39(10):111-118.doi: 10.11883/bzycj-2019-0233. Xu H, Du HM, Fan JB, et al. Research on low-frequency characteristics and compensation method of shock wave test system[J]. Explosion and Shock,2019,39(10): 111-118. doi: 10.11883/bzycj-2019-0233. [3] 孙传猛,裴东兴,陈嘉欣,许瑞嘉,崔春生,高群昌.基于深度学习的爆炸冲击波信号重构模型[J].计测技术,2022,42(02):57-67.doi: 10.11823/j.issn.1674-5795.2022.02. 07 Sun CM, Pei DX, Chen JX, et al. Deep learning-based reconstruction model for blast shockwave signals[J]. Measurement Technology,2022,42(02):57-67. doi: 10.11823/ j.issn.1674-5795.2022.02.07 [4] 李同宇,张建中.地震射线追踪的线性走时扰动插值法[J].石油地球物理勘探,2018,53(06):1165-1174+1110-1111.doi: 10.13810/j.cnki.issn.1000-7210.2018.06.006 Li TY, Zhang JC. Linear walk-time perturbation interpolation method for seismic ray tracing[J]. Petroleum Geophysical Exploration,2018,53(06):1165-1174+1110-1111. doi: 10.13810 /j.cnki.issn.1000-7210.2018.06.006 [5] Vidale J E . Finite-difference calculation of traveltimes in three dimensions[J]. Geophysics, 1990, 55(5):521-526. [6] Zhang H , Sun X , Qi X , et al. A Modified LTI ray tracing algorithm in Lamb wave tomography[C]// International Conference on Audio. IEEE, 2008.doi: 10.1109/ICALIP. 2008.4590287 [7] 郭亚丽,韩焱,王黎明.基于广义逆算法的冲击波超压场重建方法[J].爆炸与冲击,2014,34(06):764-768.doi: 10.11883/1001-1455(2014)06-0764-05. Guo YL, Han Y, Wang LM. A reconstruction method of shock wave overpressure field based on generalized inverse algorithm[J]. Explosion and Shock,2014,34 (06):764-768. doi:10.11883/1001-1455(2014)06-0764-05. [8] 李寒宇. 矩阵加权广义逆与加权极分解研究[D].重庆大学,2009.doi: 10.7666/d.y1666340. Li Hanyu. Research on matrix weighted generalized inverse and weighted polar decomposition [D]. Chongqing University, 2009. doi: 10.7666/d.y1666340. [9] 尧礼辉. 广义逆矩阵计算及在矩阵方程中应用的研究[D].解放军信息工程大学,2008. Yao, LiHui. Research on generalized inverse matrix calculation and its application in matrix equations[D]. PLA Information Engineering University, 2008. [10] 王振宇,刘国华,梁国钱.基于广义逆的层析成像反演方法研究[J].浙江大学学报(工学版),2005(01):2-6.doi: 10.3785/j.issn.1008-973X.2005.01.001. Wang Z.Y., Liu G.H., Liang G.Q.. Study on the inversion method of laminar imaging based on generalized inverse[J]. Journal of Zhejiang University (Engineering Edition),2005(01):2-6. doi: 10.3785/j.issn.1008-973X. 2005.01.001. [11] 李平,李平,许厚泽.地球物理抗差估计和广义逆方法[J].地球物理学报,2000(02):232-240.doi: 10.3321/j.issn:0001-5733. 2000.02.011. Li P, Li P, Xu HZ. Geophysical resistance estimation and generalized inverse methods[J]. Journal of Geophysics, 2000(02):232-240. doi: 10.3321/j.issn:0001-5733.2000.02. 011. [12] 杨志,张志杰,夏永乐.基于B样条插值拟合的冲击波超压场重建[J].科学技术与工程,2016,16(07):236-240.doi: 10.3969/j.issn.1671-1815.2016.07.040. Yang Z, Zhang CJ, Xia YL. Shock wave overpressure field reconstruction based on B-sample interpolation fitting[J]. Science Technology and Engineering,2016,16(07): 236-240. doi: 10.3969/j.issn.1671-1815.2016.07.040. [13] 温佩芝,宁如花,黄锦芳.基于参数优化的多层次单元划分曲面重建[J].计算机应用,2011,31(07):1811-1814.doi: 10.3724/SP.J.1087.2011.01811. Wen PZ, Ning RH, Huang JF. Multilevel cell division surface reconstruction based on parameter optimization[J]. Computer Applications,2011,31(07):1811-1814. doi: 10.3724/SP.J.1087. 2011.01811. [14] 赵化彬,张志杰.基于非均匀有理B样条“蛛网”插值的冲击波压力场重建方法[J].科学技术与工程,2017,17(18):258-264.doi: 10.3969/j.issn.1671-1815.2017. 18. 041. Zhao, HB, Zhang, CJ. Shock wave pressure field reconstruction method based on non-uniform rational B-sample "spider web" interpolation[J]. Science, Technology and Engineering,2017, 17(18):258-264. doi: 10.3969/j.issn. 1671-1815.2017.18.041. [15] 白苗苗,郭亚丽,王黎明.基于EM算法的爆炸超压场重建技术[J].弹箭与制导学报,2014,34(03):187-190.doi: 10.15892/j.cnki. djzdxb.2014.03.003. Bai MM, Guo YL, Wang LM. Explosive overpressure field reconstruction technique based on EM algorithm[J]. Journal of Ballistic Arrows and Guidance,2014,34(03): 187-190. doi: 10.15892/j.cnki.djzdxb.2014.03.003. [16] 张晓光. 冲击波超压场重建技术研究[D].中北大学,2018. Zhang Xiaoguang. Research on shock wave overpressure field reconstruction technique[D]. North Central University,2018. (in Chinese) [17] 姚悦. 近地静爆冲击波场重建技术研究[D].中北大学,2019. Yao Yue. Research on near-ground static blast shock wave field reconstruction technology[D]. North Central University,2019. [18] 曹海燕,叶震宇.基于压缩感知理论的大规模MIMO系统下行信道估计中的导频优化理论分析与算法设计[J/OL].物理学报:1-21[2022-01-14].doi: 10.7498/aps.71.20211504. Cao, HY, Ye, ZY. Theoretical analysis and algorithm design of optimized pilot for downlink channel estimation in massive MIMO systems based on compressed sensing [J/OL]. Acta Physica Sinica:1-21[2022-01-14]. doi: 10.7498/aps.71.20211504. [19] 张帅,杨润海,高尔根.基于压缩感知的信号重建方法及在气枪震源信号处理中的应用[J].地震工程学报,2021,43(02):322-330.doi: 10.3969/j.issn.1000-0844.2021. 02.322. Zhang S, Yang RH, Gao EG. Compression-aware signal reconstruction method based on compressional perception and its application in airgun seismic signal processing[J]. Journal of Earthquake Engineering,2021,43(02):322-330. doi: 10.3969/j.issn.1000-0844.2021.02.322. [20] Xu G , Xu Z . Compressed Sensing Matrices from Fourier Matrices[J]. 2013. doi: 10.1109/TIT.2014.2375259. [21] Eldar Y C , Kutyniok G . Compressed Sensing: Theory and Applications[J]. Cambridge University Press, 2012. [22] David L . Donoho. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4):1289-1306.doi: 10.1109/TIT.2006.871582 [23] 王强,张培林,王怀光等.基于优化分类的机械振动信号压缩感知[J].振动与冲击,2018,37(14):86-93. Wang Q, Zhang PL, Wang HG, et al. Compressed sensing algorithm for machinery vibration signals based on optimal classification[J]. Journal of Vibration and Shock, 2018,37(14): 86-93. [24] 姚舜才, 孙传猛. 机器学习基础教程[M].西安电子科技大学出版社, 2020. Yao SC, Sun CM. Machine Learning Fundamentals Tutorial [M]. Xi'an University of Electronic Science and Technology Press, 2020. [25] 赵婷婷,韩雅杰,杨梦楠,等.基于机器学习的时序数据预测方法研究综述[J/OL].天津科技大学学报:1-9[2021-10-17]. doi: 10.13364/j.issn.1672-6510.20200203. Zhao TT, Han YJ, Yang MN, et al. Review of prediction methods of time series data based on machine learning [J/OL]. Journal of Tianjin University of Science and Technology:1-9 [2021-10-17]. doi: 10.13364/j.issn. 1672-6510.20200203. [26] 郭俊锋,王茁.一种基于多测量向量模型的机械振动信号联合稀疏重构方法[J].振动与冲击,2021,40(01):254-263. Guo JF, Wang Z. A joint sparse reconstruction method for mechanical vibration signals based on multi-measurement vector model [J]. Journal of Vibration and Shock, 2021,40(01): 254-263. [27] 王旭磊. 基于GAN和GRU的时间序列预测和填补方法研究[D].西安理工大学,2021. doi: 10.27398/d.cnki.gxalu. 2021.000898. Wang Xulei. Research on time series prediction and filling methods based on GAN and GRU [D]. Xi'an University of Technology,2021. doi: 10.27398/d.cnki.gxalu.2021.000898. [28] 孙传猛,王燕平,王冲等.融合改进YOLOv3与三次样条插值的煤岩界面识别方法[J].采矿与岩层控制工程学报,2022,4(01):81-90. Sun CM, Wang YP, Wang C, et al. A coal-rock interface identification method incorporating improved YOLOv3 and cubic spline interpolation[J]. Journal of Mining and Seam Control Engineering,2022,4(01):81-90. [29] 王鑫. 基于生成对抗网络模型的数据填补技术研究[D].大连理工大学,2021.doi: 10.26991/d.cnki.gdllu.2021.003519. Wang Xin. Research on data filling techniques based on generative adversarial network model [D]. Dalian University of Technology,2021. doi: 10.26991/d.cnki.gdllu.2021.003519. [30] 罗永洪. 基于生成对抗网络的时序数据缺失值填充算法研究[D].南开大学,2019. Luo YH. Research on missing value filling algorithm for temporal data based on generative adversarial network[D]. Nankai University, 2019. [31] 豆佳敏. 基于深度学习的冲击波信号压缩感知方法[D].长春理工大学,2021. Dou JiaMin. Deep learning-based shock wave signal compression perception method[D]. Changchun University of Science and Technology,2021. [32] CYBENKO G. Approximations by superpositions of a sigmoidal function[J]. Mathematics of Contrl, Signals and Systems, 1989, 2: 183-192. doi: doi.org/10.1007/bf02551274. [33] HORNIK K, STINCHCOMBE M, WHITE H. Multilayer feedforward networks are universal approximators[J]. Neural networks, 1989, 2(5): 359-366. doi: doi.org/10.1016/ 0893-6080(89)90020-8. [34] He K , Zhang X , Ren S , et al. Deep Residual Learning for Image Recognition[C]. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016, pp. 770-778.doi: 10.1109/CVPR.2016.90. [35] 江知航,王艳霞,颜家均等.基于BILSTM的棉花价格预测建模与分析[J].中国农机化学报,2021,42(08):151-160. doi: 10.13733/j.jcam.issn.2095-5553.2021.08.21. Jiang ZH, Wang YX, Yan JJ et al. Modeling and analysis of cotton price prediction based on BILSTM[J]. Chinese Journal of Agricultural Chemistry,2021, 42(08):151-160. doi: 10.13733/j.jcam.issn.2095-5553.2021.08.21.
PDF(3700 KB)

294

Accesses

0

Citation

Detail

Sections
Recommended

/