神经网络在红外热波无损检测中的应用  

Application of neural network in infrared thermal wave nondestructive detection

作　　者：马静[1] 

机构地区：[1]内蒙古工业大学乌海学院,内蒙古乌海市016000

出　　处：《激光杂志》2016年第8期79-81,共3页Laser Journal

基　　金：内蒙古自治区高等学校科学研究项目(NJZC363)

摘　　要：为了提高红外热波无损检测的定量识别精度,提出一种改进神经网络的红外热波无损检测方法。该方法将最佳检测时间、最佳温差等因素作为红外热波无损检测定量识别的输入,表现缺陷的深度和直径作为BP神经网络的期望输出,通过搜索优化算法优化BP神经网络对输入和输出之间关系进行拟合,并进行了多次测试实验,结果表明,本文方法可以降低红外热波无损检测定量识别的误差。In order to improve quantitative recognition accuracy of infrared thermal wave nondestructive detection, an infrared thermal wave nondestructive detection method based on improved neural network is proposed. The best de- tection and differential temperature is taken as input of infrared thermal wave nondestructive quantitative identification detection while depth and diameter of the defect is taken as expected output of the BP neural network, and the relation- ship between input and output is simulated by BP neural network which is optimized by seeker optimization algorithm, the test experiment is carried out. The results show that the proposed method can reduce the quantitative recognition error of infrared thermal wave nondestructive detection.

关 键 词：红外技术 BP神经网络 热波无损检测 分类与识别 

分 类 号：TN219[电子电信—物理电子学]