基于遗传算法的亚表面缺陷热波散射反问题  

Inverse Problem of Thermal Wave Scattering from Subsurface Defects Based on Genetic Algorithm

作　　者：马晓波[1] 吴修涵 MA Xiaobo;WU Xiuhan(School of Mechanical Engineering,Tongji University,Shanghai 201804,China)

机构地区：[1]同济大学机械与能源工程学院

出　　处：《红外技术》2019年第11期1077-1083,共7页Infrared Technology

基　　金：国家自然科学基金（51276129）

摘　　要：红外热波检测技术被应用于零部件或设备内部缺陷及故障的检测与定量识别。本文建立了半无限材料中圆柱绝热缺陷在超短激光脉冲辐照下的非傅里叶热传导波动模型,在采用波函数展开法和镜像法研究热波散射正问题的基础上,运用遗传算法对亚表面缺陷深度进行反演,实际测量温度取热波散射正问题计算得到的材料表面温度,并且分析讨论了测温点数目、入射频率等因素对反演结果的影响。数值实验结果证明了该方法的有效性。与L-M法和共轭梯度法相比,该方法在反演的精度及速度上都有较为良好的表现。Infrared thermal wave detection technology is applied to detect and quantitatively identify the internal defects and faults of parts or equipment. A non-Fourier heat conduction wave model of a cylindrical insulation defect in a semi-infinite material under ultra short laser pulses is established in this paper. On the basis of studying the direct problem of the model by using the wave function expansion and image methods, the inverse problem is studied by a genetic algorithm. The author uses the surface temperature of the material calculated from the direct problem as the actual measurement temperature. In this paper, the depth of a subsurface defect is inverted, and the influence of a series of factors such as the number of temperature measuring points and incident frequency on the inversion results, is analyzed and discussed. In addition, numerical experimental results certify the effectiveness of the method. Compared with the Levenberg-Marquardt and conjugate gradient methods, this method has better performance in terms of both accuracy and speed of inversion.

关 键 词：超短激光脉冲 非傅里叶热传导 导热反问题 遗传算法 

分 类 号：TK124[动力工程及工程热物理—工程热物理]