基于人工神经网络的钢铁件淬火硬度电磁无损检测

Electromagnetic Non-destructive Test for Quenched Hardness of Iron and Steel Parts Based on ANN

利用DWY-1型电磁无损检测仪,采用改进的Gram-Schm idt方法优化的RBF(径向基函数网络)人工神经网络,实现了钢铁件淬火硬度的实时在线无损检测。结果表明,淬火硬度的检测精度、网络的收敛速度能满足生产实际的需要。

By means of the new type of electromagnetic nondestructive measurement instrument （ DWY-1 ） , using the RBF （ Radial Basis Function） neural network optimized by modified Gram-Schmidt method,the real-time on line non-destructive detect for hardness of steel parts was realized. The results show that the detect accuracy and the convergence rate of network can meet the requirement of practical production.