Moving force identification based on modified preconditioned conjugate gradient method

This paper develops a modified preconditioned conjugate gradient (M-PCG) method for moving force identification (MFI) by improving the conjugate gradient (CG) and preconditioned conjugate gradient (PCG) methods with a modified Gram-Schmidt algorithm. The method aims to obtain more accurate and more efficient identification results from the responses of bridge deck caused by vehicles passing by, which are known to be sensitive to ill-posed problems that exist in the inverse problem. A simply supported beam model with biaxial time-varying forces is used to generate numerical simulations with various analysis scenarios to assess the effectiveness of the method. Evaluation results show that regularization matrix L and number of iterations j are very important influence factors to identification accuracy and noise immunity of M-PCG. Compared with the conventional counterpart SVD embedded in the time domain method (TDM) and the standard form of CG, the M-PCG with proper regularization matrix has many advantages such as better adaptability and more robust to ill-posed problems. More importantly, it is shown that the average optimal numbers of iterations of M-PCG can be reduced by more than 70% compared with PCG and this apparently makes M-PCG a preferred choice for field MFI applications.

[1]  E. Schmidt.  Über die Auflösung linearer Gleichungen mit Unendlich vielen unbekannten , 1908 .

[2]  Tommy H.T. Chan,et al.  Moving Force Identification based on the Frequency-Time Domain Method , 2003 .

[3]  Zhen Chen,et al.  A truncated generalized singular value decomposition algorithm for moving force identification with ill-posed problems , 2017 .

[4]  Tospol Pinkaew,et al.  Identification of vehicle axle loads from bridge responses using updated static component technique , 2006 .

[5]  Robert Plato,et al.  The conjugate gradient method for linear ill-posed problems with operator perturbations , 1999, Numerical Algorithms.

[6]  Eugene J. O'Brien,et al.  Adaptation of Cross Entropy optimisation to a dynamic Bridge WIM calibration problem , 2012 .

[7]  Tommy H.T. Chan,et al.  Moving axle load from multi-span continuous bridge: Laboratory study , 2006 .

[8]  S. S. Law,et al.  Regularization in moving force identification , 2001 .

[9]  Michele Benzi,et al.  A Sparse Approximate Inverse Preconditioner for the Conjugate Gradient Method , 1996, SIAM J. Sci. Comput..

[10]  H. Benaroya,et al.  Review of force reconstruction techniques , 2014 .

[11]  Siu-Seong Law,et al.  Moving Force Identification Studies I: Theory , 2001 .

[12]  Siu-Seong Law,et al.  Moving Force Identification Studies II: Comparative Studies , 2001 .

[13]  Sq Q. Wu,et al.  Moving force identification based on stochastic finite element model , 2010 .

[14]  Achiya Dax,et al.  A modified Gram–Schmidt algorithm with iterative orthogonalization and column pivoting , 2000 .

[15]  S. S. Law,et al.  Identification of moving interaction forces with incomplete velocity information. , 2003 .

[16]  G. Golub,et al.  Block Preconditioning for the Conjugate Gradient Method , 1985 .

[17]  Chudong Pan,et al.  Moving force identification based on redundant concatenated dictionary and weighted l 1 -norm regularization , 2018 .

[18]  J. Rice.  Experiments on Gram-Schmidt orthogonalization , 1966 .

[19]  Maria Q. Feng,et al.  Simultaneous identification of bridge structural parameters and vehicle loads , 2015 .

[20]  Zhu Junhua.  Moving Load Identification Based on Preconditioned Conjugate Gradient Method , 2008 .

[21]  M. Hanke.  Regularization with differential operators : an iterative approach , 1992 .

[22]  Eugene J. O'Brien,et al.  A general solution to the identification of moving vehicle forces on a bridge , 2008 .

[23]  Tommy H.T. Chan,et al.  Dynamic wheel loads from bridge strains , 1988 .

[24]  C. Jiang,et al.  A novel computational inverse technique for load identification using the shape function method of moving least square fitting , 2014 .

[25]  Yu Ling.  Effect of preconditioned matrix on PCGM used for identification of dynamic axle loads on bridge , 2009 .

[26]  Tommy H.T. Chan,et al.  Recent research on identification of moving loads on bridges , 2007 .

[27]  Tommy H.T. Chan,et al.  Theoretical study of moving force identification on continuous bridges , 2006 .

[28]  G. Meurant,et al.  The Lanczos and conjugate gradient algorithms in finite precision arithmetic , 2006, Acta Numerica.

[29]  Tommy H.T. Chan,et al.  Moving force identification - A frequency and time domains analysis , 1999 .

[30]  Axel Ruhe.  Numerical aspects of gram-schmidt orthogonalization of vectors , 1983 .

[31]  Xinqun Zhu,et al.  IDENTIFICATION OF VEHICLE AXLE LOADS FROM BRIDGE DYNAMIC RESPONSES , 2000 .

[32]  Arturo González,et al.  Wavelet domain analysis for identification of vehicle axles from bridge measurements , 2006 .

[33]  Chen Zhen.  EXPERIMENTAL STUDY ON ILL-CONDITIONING PROBLEM IN IDENTIFICATION OF MOVING VEHICLE LOADS ON BRIDGES , 2007 .

[34]  C. S. Manohar,et al.  Substructuring tools for probabilistic analysis of instrumented nonlinear moving oscillator-beam systems , 2017 .

[35]  Xingxing Jiang,et al.  Reconstruction of distributed dynamic loads on an Euler beam via mode-selection and consistent spatial expression , 2008 .

[36]  M. Hestenes,et al.  Methods of conjugate gradients for solving linear systems , 1952 .

[37]  Tommy H.T. Chan,et al.  A MOM-based algorithm for moving force identification: Part I . Theory and numerical simulation , 2008 .

[38]  S. S. Law,et al.  An interpretive method for moving force identification , 1999 .

[39]  Radosław Pytlak,et al.  Conjugate Gradient Algorithms in Nonconvex Optimization , 2008 .

[40]  Tommy H.T. Chan,et al.  Moving force identification: A time domain method , 1997 .

[41]  S. S. Law,et al.  Moving load identification on multi-span continuous bridges with elastic bearings , 2006 .