|
|
| |
| 自适应窗长改进Sage-Husa卡尔曼滤波的结构响应重构 |
| Structural response reconstruction of Adaptive window length improved Sage-Husa Kalman filter |
| 投稿时间：2023-09-26 修订日期：2023-11-04 |
| DOI： |
| 中文关键词: 卡尔曼滤波 结构响应重构 噪声估计 UD矩阵分解 |
| 英文关键词:Kalman filter Structural response reconstruction Noise estimation UD matrix decomposition |
| 基金项目:国家自然科学基金项目(62161018) |
|
| 摘要点击次数: 157 |
| 全文下载次数: 0 |
| 中文摘要: |
| 针对固定窗口估计噪声方差的卡尔曼滤波方法在进行结构响应重构时窗口长度由经验选取导致重构精度降低的问题，提出了一种自适应窗长改进Sage-Husa卡尔曼滤波(Improved Sage-Husa Kalman filter, ISHKF)算法的结构响应重构方法。首先，对卡尔曼滤波先验误差协方差矩阵应用UD分解以保证其正定性，根据测量先验误差自适应调整窗口长度，再应用Sage-Husa时变噪声估计器，实现测量噪声和系统噪声的同步调节。然后，根据已知的部分加速度响应，结合模态叠加法重构加速度、速度和位移响应。最后通过某动车组C型支架数值仿真和简支梁试验来验证方法的可行性。结果表明所提方法可实现自适应窗长未知噪声方差估计下的结构响应重构，与移动窗卡尔曼滤波方法相比，重构精度更高。 |
| 英文摘要: |
| In order to solve the problem that the precision of structural response reconstruction is reduced due to the empirical selection of window length in the fixed window estimation noise variance Kalman filter method, a structural response reconstruction method based on adaptive window length improved Sage-Husa Kalman filter (ISHKF) is proposed. Firstly, the UD decomposition is applied to the Kalman filter prior error covariance matrix to ensure its positive definiteness, and the window length is adaptively adjusted according to the measurement prior error, and then the Sage-Husa time-varying noise estimator is used to realize the synchronous adjustment of measurement noise and system noise. Secondly, according to the known partial acceleration response, combined with the modal superposition method, the acceleration, velocity and displacement response are reconstructed. Finally, the feasibility of the method is verified by numerical simulation of C-bracket and simply supported beam test of an EMU. The results show that the proposed method can realize the structural response reconstruction under adaptive window length unknown noise variance estimation, and the reconstruction accuracy is higher than the moving window Kalman filter method. |
| 查看/发表评论 下载PDF阅读器 |
