一种改进的人工神经网络学习算法及其在超声检测中的应用

An improved learning algorithm for artificial neural network and its application in ultrasonic testing

本文用多层感知器 (MLP)与误差反向传播算法 (errorback propagationalgorithm)构造训练人工神经网络 ,提出了新的误差反向传播改进算法。试验结果表明 ,改进的BP算法收敛速度较之常规BP算法明显加快 ,因而在工业现场的超声检测领域有广阔应用前景。

The model of multilayer perceptron (MLP) and back-propagation (BP) training algorithm in artificial neural network are employed in this paper. New ideas are proposed to improve learning algorithm in aspects of learning rate for BP training. Experiment results are also presented to demonstrate the effect of improvement, which has a widely applied future for ultrasonic testing in industry.