Online Monitoring of Welding Status Based on a DBN Model During Laser Welding

Abstract In this research, an auxiliary illumination visual sensor system, an ultraviolet/visible (UVV) band visual sensor system (with a wavelength less than 780 nm), a spectrometer, and a photodiode are employed to capture insights into the high-power disc laser welding process. The features of the visible optical light signal and the reflected laser light signal are extracted by decomposing the original signal captured by the photodiode via the wavelet packet decomposition (WPD) method. The captured signals of the spectrometer mainly have a wavelength of 400–900 nm, and are divided into 25 sub-bands to extract the spectrum features by statistical methods. The features of the plume and spatters are acquired by images captured by the UVV visual sensor system, and the features of the keyhole are extracted from images captured by the auxiliary illumination visual sensor system. Based on these real-time quantized features of the welding process, a deep belief network (DBN) is established to monitor the welding status. A genetic algorithm is applied to optimize the parameters of the proposed DBN model. The established DBN model shows higher accuracy and robustness in monitoring welding status in comparison with a traditional back-propagation neural network (BPNN) model. The effectiveness and generalization ability of the proposed DBN are validated by three additional experiments with different welding parameters.

[1]  Zhen-Hua Ling,et al.  DBN-based Spectral Feature Representation for Statistical Parametric Speech Synthesis , 2016, IEEE Signal Processing Letters.

[2]  Radovan Kovacevic,et al.  Real-time monitoring of laser welding of galvanized high strength steel in lap joint configuration , 2012 .

[3]  S.-K. Lee,et al.  Sound quality evaluation of the impact noise induced by road courses having an impact bar and speed bumps in a passenger car , 2010 .

[4]  Sang-Kwon Lee,et al.  Improvement of impact noise in a passenger car utilizing sound metric based on wavelet transform , 2010 .

[5]  Yanxi Zhang,et al.  Effect of joint gap on bead formation in laser butt welding of stainless steel , 2017 .

[6]  Deyong You,et al.  Seam Tracking Monitoring Based on Adaptive Kalman Filter Embedded Elman Neural Network During High-Power Fiber Laser Welding , 2012, IEEE Transactions on Industrial Electronics.

[7]  Yanxi Zhang,et al.  Analysis of characteristics of molten pool using cast shadow during high-power disk laser welding , 2014 .

[8]  Hai Liu,et al.  Sound quality prediction for engine-radiated noise , 2015 .

[9]  Teik C. Lim,et al.  Evaluation of vehicle interior sound quality using a continuous restricted Boltzmann machine-based DBN , 2017 .

[10]  Yanxi Zhang,et al.  Weld appearance prediction with BP neural network improved by genetic algorithm during disk laser welding , 2015 .

[11]  Susumu Tsukamoto,et al.  High speed imaging technique Part 2 – High speed imaging of power beam welding phenomena , 2011 .

[12]  Geoffrey E. Hinton,et al.  Acoustic Modeling Using Deep Belief Networks , 2012, IEEE Transactions on Audio, Speech, and Language Processing.

[13]  Seiji Katayama,et al.  Relationship between spatter formation and dynamic molten pool during high-power deep-penetration laser welding , 2014 .

[14]  Yan Song Wang,et al.  A sound quality model for objective synthesis evaluation of vehicle interior noise based on artificial neural network , 2014 .

[15]  X. Shao,et al.  3D transient multiphase model for keyhole, vapor plume, and weld pool dynamics in laser welding including the ambient pressure effect , 2015 .

[16]  Zou Cairong,et al.  A Novel DBN Feature Fusion Model for Cross-Corpus Speech Emotion Recognition , 2016 .

[17]  T. DebRoy,et al.  Mechanisms of Spiking and Humping in Keyhole Welding , 2012, IEEE Transactions on Components, Packaging and Manufacturing Technology.

[18]  S. Na,et al.  Theoretical analysis of keyhole dynamics in polarized laser drilling , 2007 .

[19]  Ping Zhang,et al.  A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process , 2012 .

[20]  Yigang He,et al.  Analog Circuit Incipient Fault Diagnosis Method Using DBN Based Features Extraction , 2018, IEEE Access.

[21]  Giovanni Muscato,et al.  An Arc Welding Robot Control for a Shaped Metal Deposition Plant: Modular Software Interface and Sensors , 2011, IEEE Transactions on Industrial Electronics.

[22]  De Xu,et al.  A Vision-Based Self-Tuning Fuzzy Controller for Fillet Weld Seam Tracking , 2011, IEEE/ASME Transactions on Mechatronics.

[23]  X. Shao,et al.  Dynamics of vapor plume in transient keyhole during laser welding of stainless steel: Local evaporation, plume swing and gas entrapment into porosity , 2016 .

[24]  R. Kovacevic,et al.  High power laser welding of thick steel plates in a horizontal butt joint configuration , 2016 .

[25]  Steven X. Ding,et al.  Real-Time Implementation of Fault-Tolerant Control Systems With Performance Optimization , 2014, IEEE Transactions on Industrial Electronics.

[26]  Honglak Lee,et al.  Learning hierarchical representations for face verification with convolutional deep belief networks , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[27]  Xinran Zhang,et al.  A Novel DBN Feature Fusion Model for Cross-Corpus Speech Emotion Recognition , 2016, J. Electr. Comput. Eng..

[28]  Yi Zhang,et al.  Correlation analysis of penetration based on keyhole and plasma plume in laser welding , 2018, Journal of Materials Processing Technology.

[29]  Xiao-Lei Zhang,et al.  Deep Belief Networks Based Voice Activity Detection , 2013, IEEE Transactions on Audio, Speech, and Language Processing.

[30]  X. Shao,et al.  Efficient multiple time scale method for modeling compressible vapor plume dynamics inside transient keyhole during fiber laser welding , 2016 .