主成分分析结合神经网络技术在焊接质量控制中的应用

Principal component analysis based artificial neural networks for arc welding quality control

介绍了主成分分析方法及人工神经网络技术在相关因素分析和质量控制的建模与估计中的应用。以大电流MAG焊熔宽控制为例 ,通过对 6个焊接过程参数进行主成分分析 ,提取出影响熔宽的 4个主要因素。讨论了提取的主成分与原始过程参数间的关系。以主成分得分作为新的训练样本集 ,送入神经网络进行计算。结果表明 ,基于主成分分析的神经网络无论在收敛速度 ,还是在训练精度上 ,都远远优于基本BP神经网络。

The application of principal component analysis (PCA) and artificial neural networks (ANN) to the multivariate statistical analysis and quality control was introduced. The pool width control of MAG weld with high current was taken as an example. Through the PCA of 6 welding parameters, 4 main factors were extracted. The relationship between main factors and original parameters was discussed. The PCA values were taken as the new training sample set and the output results indicated both the convergent speed and the training accuracy of PCA-based ANN were much better than those of basic BP ANN.