抄録
Simulation analysis is an effective method for assessing the safety of a building after an earthquake. This kind of analysis is effective for evaluating foundations or piles, whose safety cannot be confirmed by a visual check.
During such analyses, the input motion should be set to reproduce observed waves at observed points if it exists. In the case of a single target wave the modal iterative error correction (MIEC) method can be used to set the input motion even if the analysis model exhibits nonlinear behavior. However, this method is not applicable to multi-point observed waves.
In this study, the MIEC method was modified to be applicable to multi-point target waves. The precision of the method was confirmed experimentally.
The findings of this study were as follows:
(1) A new time domain inversion method was developed for multi-point records. The proposed method improves upon the method for single-point records. The proposed method increases the size of the output vector according to the number of target points. Then, the perturbation impulse matrix becomes a rectangular matrix from a square matrix. The main mode of the general inverse of this rectangular matrix is used to calculate the input correction amount. These corrections are then repeated until the error becomes sufficiently small.
(2) The proposed method was applied to a sample problem to confirm its accuracy. An identified input motion was confirmed to reproduce target output motions at target points of the elastoplastic soil model. Moreover, using more acceleration records increased the convergence speed.
The findings of this study confirmed that the proposed method can be applied to inversion. However, the procedure needs to be further improved because a large amount of calculation time is needed even when multi-point records are used.
