• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.26 no.2
• /
• pp.99-112
• /
• 2013

This paper presents a subspace system identification for estimating the stiffness matrix and flexural rigidities of a shear building. System matrices are estimated by LQ decomposition and singular value decomposition from an input-output Hankel matrix. The estimated system matrices are converted into a real coordinate through similarity transformation, and the stiffness matrix is estimated from the system matrices. The accuracy and the stability of an estimated stiffness matrix depend on the size of the associated Hankel matrix. The estimation error curve of the stiffness matrix is obtained with respect to the size of a Hankel matrix using a prior finite element model of a shear building. The sizes of the Hankel matrix, which are consistent with a target accuracy level, are chosen through this curve. Among these candidate sizes of the Hankel matrix, more proper one can be determined considering the computational cost of subspace identification. The stiffness matrix and flexural rigidities are estimated using the Hankel matrix with the candidate sizes. The validity of the proposed method is demonstrated through the numerical example of a five-story shear building model with and without damage.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.38 no.12
• /
• pp.56-65
• /
• 2001

This paper proposes an algorithm for the capacitance extraction of general 3-dimensional conductors in an ideal uniform dielectric that uses a high-order quadrature approximation method combined with the typical first-order collocation method to enhance the accuracy and adopts an efficient matrix-vector product algorithm for the model-order reduction to achieve efficiency. The proposed method enhances the accuracy using the quadrature method for interconnects containing corners and vias that concentrate the charge density. It also achieves the efficiency by reducing the model order using the fact that large parts of system matrices are of numerically low rank. This technique combines an SVD-based algorithm for the compression of rank-deficient matrices and Gram-Schmidt algorithm of a Krylov-subspace iterative technique for the rapid multiplication of matrices. It is shown through the performance evaluation procedure that the combination of these two techniques leads to a more efficient algorithm than Gaussian elimination or other standard iterative schemes within a given error tolerance.

• Journal of the Korean Society for Precision Engineering
• /
• v.17 no.8
• /
• pp.78-86
• /
• 2000

This study identified a linear time-invariant mathematical model of each stand of a five-stand tandem cold mill. Two model identification methods are applied to construct a linear model of each stand of the tandem cold mill. For the model identification the input-output data that have interstand interference property in tandem cold rolling are obtained from a nonlinear simulator of the tandem cold mill. And a linear model of each stand is identified with N4SD(numerical algorithms for subspace state space system identification) method based on a state-space model and Least Square algorithm based on a transfer function. Furthermore a modeling error of the tandem cold mill is quantitatively analyzed from a maximum singular value plot of error function between an identified nominal model and uncertain model. In conclusion the comparison of the output signals between the existing Taylor linearized model the identified linear model and the nonlinear model of the tandem cold mill shows the accuracy and the applicability of the proposed identified model.