研究課題/領域番号 |
21K14284
|
研究機関 | 京都大学 |
研究代表者 |
郭 佳 京都大学, 農学研究科, 准教授 (50868081)
|
研究期間 (年度) |
2021-04-01 – 2024-03-31
|
キーワード | Deep neural network / Force reconstruction / Linear multistep method / Data-driven approach |
研究実績の概要 |
Despite great progress in identifying the hysteretic forces in structures, less satisfactory hysteretic behaviors with large errors for the nonlinear component reconstruction might still be identified, due to the ill-posedness of the inverse problem. This year, we improved the Kalman filter based restoring force reconstruction method through incorporating deep neural networks into the classical numerical integration method by using a hybridized integration time-stepper. We proposed to use residual network to generally identify the nonlinear behaviors of the system. Compared to the previous method, the newly developed Physics-DNN hybridized integration time-stepping scheme provides stable solution in both of the hysteresis identification and the corresponding structural dynamic analysis.
|
現在までの達成度 (区分) |
現在までの達成度 (区分)
3: やや遅れている
理由
In 2021, a real-time hysteresis identification method has been successfully established based on restoring force reconstruction and the Kalman filter. In 2022, this method is combined with the deep learning process so that a mathematical model is involved during the identification process to circumvent the ill-posedness problem. In this way, more stable identification results can be obtained. Besides, not only numerical simulations, experimental tests were also conducted to demonstrate the performance of the method, as planned. However, due to the ongoing spread of COVID-19, this method has not yet been tested using actual field measurement data. Therefore, we have decided to extend this project by one year in order to complete our final validation.
|
今後の研究の推進方策 |
In 2023, we will continue to enhance the proposed method using advanced computational techniques to make it more practical and feasible for real-world applications. Moreover, the proposed method will be further validated using newly generated numerical data from complex nonlinear structural systems or field measurements from a Structural Health Monitoring system installed in a real high-rise building, if possible. The researching outcome will be summarized and submitted to international academic journal.
|
次年度使用額が生じた理由 |
In the final fiscal year, the budget will only be allocated for expenses to article costs relating to the research (e.g. books), or registration/participation fees for academic conferences.
|