This paper proposes an inverse analysis method for identification of unknown defects and/or unknown boundary conditions of an elastic body in a steady elastodynamic state. An extended Kalman filter and the boundary element method are combined and applied to the inverse analysis method. The extended Kalman filter algorithm enables estimation of state variables of a stochastic system. It is expected that the algorithm is also applicable to analysis of other identification problems for which only noisy data are available at certain measuring points. The effectiveness of the proposed inverse analysis method is demonstrated through numerical simulation for several example problems.
抄録全体を表示