The Principal Component Analysis(PCA)is one of the multivariate statistical methods.It has the special advantage in dealing with the relationship among a large number of
parameters,excluding the minor factors and extracting the major ones.As an effective model method,the Artificial Neural Networks(ANN)had been applied in production extensively.The problem was that excessive input varies and complex
structure often made the convergence accuracy descend and the modelling effect worse.Through improving the input factors,eliminating the correlation among the inputs and simplifying the structure,the PCA-
based ANN could accelerate the learning process and increase the convergence accuracy.In this paper,the pool width
control of MAG weld with high
current was taken as an example.Through the PCA of6
welding parameters(
wire feed rate,wire extension,welding speed,gas flow,welding voltage and welding current),4main factors were extracted to form the new training samples.The output results of ANN are satisfied.
Published: April 28, 2003
More reviews about the PCA-based ANN for arc welding quality control