基于阻尼LSQR算法的混凝土无损检测技术的研究与应用
摘要
在混凝土超声无损检测中,采用CT层析成像的方法,能够呈现出混凝土内部的结构特征,更直观显示缺陷的位置、大小及类型。针对常规LSQR(Least Squares QR-factorization)迭代算法在数据误差较大时仍会发散的缺点,为了提高LSQR算法的抗噪能力,在LSQR算法迭代求解过程中施加一定的阻尼,控制减小奇异值对解的影响,这就是阻尼LSQR(Damping Least Squares QR-factorization)算法。计算机仿真结果及混凝土试件试验结果表明:阻尼LSQR算法比LSQR算法具有更好的CT重建效果和消噪能力,提高了计算的精度和图像重建质量。
引文
[1]Nolet G,著.地震层析成像及应用[M].学苑出版社,1989.
[2]杨文采,李幼铭,等编著.应用地震层析成像[M].地质出版社,1993.
[3]Martinez O,Akhnak M,Ullat L G,F Montero de Es-pinosa.A small 2D ultrasonic array for NDTapplications[J].NDT&e International,2003,36:57-63.
[4]黄联捷,李幼铭,昊如山.用于图像重建的波前法射线追踪[J].地球物理学报,1992,35:223-232.
[5]Paige C C,Saunders M A.LSQR:An algorithm forsparse linear equations and sparse least squares[J].AMC Transactions Math.Software,1982,(8):43-71.
[6]Chen Tangmin.Application of Tomography Technolo-gy Based on amping LSQR Algorithm to Casting FlawInspection[J].Transactions of China ElectrotechnicalSociety,2006,21(11):110-115.
[7]Wan X L,Xi D Y,Gao E G,et al.3-D resistivity in-version by the least-squares QR factorization methodunder improved smoothness constraint condition[J].Chinese J.Geophys(in Chinese),2005,48(1):439-444.
[2]杨文采,李幼铭,等编著.应用地震层析成像[M].地质出版社,1993.
[3]Martinez O,Akhnak M,Ullat L G,F Montero de Es-pinosa.A small 2D ultrasonic array for NDTapplications[J].NDT&e International,2003,36:57-63.
[4]黄联捷,李幼铭,昊如山.用于图像重建的波前法射线追踪[J].地球物理学报,1992,35:223-232.
[5]Paige C C,Saunders M A.LSQR:An algorithm forsparse linear equations and sparse least squares[J].AMC Transactions Math.Software,1982,(8):43-71.
[6]Chen Tangmin.Application of Tomography Technolo-gy Based on amping LSQR Algorithm to Casting FlawInspection[J].Transactions of China ElectrotechnicalSociety,2006,21(11):110-115.
[7]Wan X L,Xi D Y,Gao E G,et al.3-D resistivity in-version by the least-squares QR factorization methodunder improved smoothness constraint condition[J].Chinese J.Geophys(in Chinese),2005,48(1):439-444.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心