第五部門：応用数理問題（B)

[S05B-02] Numerical Verification of Noise Influence on Vehicle Parameter Estimation Accuracy of System Identification by On-going Monitoring

*Ryota Shin¹, Jun Inoue¹, Yukihiko Okada¹, Kyosuke Yamamoto¹ (1. University of Tsukuba)

Keywords:Vehicle response analysis, Vehicle-Bridge Interaction System, Paremeter identification

In this study, we devise a technology to simultaneously inspect vehicles, bridges, and road surfaces from vibration data of traveling vehicles at the same time. To estimate the parameters of vehicles, bridges and road surfaces by convergent computation, the shape of the objective function is important because it affects not only the accuracy but also the efficiency. As a result of numerical calculation, it was found that the objective function is always convex downward, even under the influence of measurement noises. This result suggests that the optimal solution can be searched by a simple method such as gradient descent.