Qualitative and quantitative analysis of gmaw welding fault based on mahalanobis distance

The concept of calculating Mahalanobis distance (MD) was introduced in order to describe welding faults. First, a set of weldments without any faults were generated in a number of repeated sessions in order to be used as references. The values of Mahalanobis distance obtained from the reference weldments were taken as a reference or a standard. Then, additional weldments were fashioned while artificial changing the flow rate of shielding gas and types of contact tips and simultaneously obtaining values for arc voltage and current at a rate of more than 8000 samples per second. Last, Mahalanobis distances for voltage and current values were calculated and used for qualitative and quantitative analysis with comparison to values obtained from the reference welds. The results described in this article confirm that Mahalanobis distances based on the short-term changes of welding parameters can be feasibly and effectively used for quality rating of welding faults in gas metal arc welding.

[1]  S. W. Simpson.  Through arc sensing in gas metal arc welding with signature images , 2008 .

[2]  Aapo Hyvärinen,et al.  A Fast Fixed-Point Algorithm for Independent Component Analysis , 1997, Neural Computation.

[3]  S. W. Simpson.  Signature images for arc welding fault detection , 2007 .

[4]  Young Joo Lee,et al.  Automatic generation of optimum cooling circuit for large injection molded parts , 2010 .

[5]  M. Ushio,et al.  Measurement and theoretical investigation of arc sensor sensitivity in dynamic state during gas metal arc welding , 1997 .

[6]  Rajesh Jugulum,et al.  The Mahalanobis-Taguchi strategy : a pattern technology system , 2002 .

[7]  X. Li,et al.  Parametric approach to positional fault detection in short arc welding , 2009 .

[8]  Alvin M. Strauss,et al.  Statistical process control application to weld process , 1997 .

[9]  Chuansong Wu,et al.  Real-time sensing and monitoring in robotic gas metal arc welding , 2006 .

[10]  Masayuki Itagaki,et al.  Diagnosis of quality of fresh water for carbon steel corrosion by Mahalanobis distance , 2007 .

[11]  Aapo Hyvärinen,et al.  Fast and robust fixed-point algorithms for independent component analysis , 1999, IEEE Trans. Neural Networks.

[12]  K. Luksa,et al.  Influence of weld imperfection on short circuit GMA welding arc stability , 2006 .

[13]  S. W. Simpson.  Statistics of signature images for arc welding fault detection , 2007 .

[14]  Jhareswar Maiti,et al.  Development of a hybrid methodology for dimensionality reduction in Mahalanobis-Taguchi system using Mahalanobis distance and binary particle swarm optimization , 2010, Expert Syst. Appl..