基于IAAKF算法的结构激励识别与响应重构  

Structural excitation identification and response reconstruction based on IAAKF algorithm

作　　者：殷红[1] 丁怡渊 彭珍瑞[1] 李鑫煜 YIN Hong;DING Yiyuan;PENG Zhenrui;LI Xinyu(School of Mechanical Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]兰州交通大学机电工程学院,兰州730070

出　　处：《振动与冲击》2024年第15期302-310,共9页Journal of Vibration and Shock

基　　金：国家自然科学基金项目(62161018)。

摘　　要：针对传统卡尔曼滤波(Kalman filter,KF)算法实际应用于响应重构时,需要已知结构外部激励并预设先验恒定噪声方差的问题,提出了一种基于IAAKF(innovation-based adaptive augmented Kalman filter)算法的结构激励识别与响应重构方法。首先,基于增广状态空间模型将外部激励向量与状态向量联合构成增广状态向量,并根据增广新息统计特性实时自适应地调整卡尔曼滤波增益和状态估计误差协方差矩阵;其次,仅借助加速度传感器根据模态法来识别锤击激励,并且重构出加速度、速度、位移以及应变响应等数据;最后,对起重机桁架和简支梁分别进行数值模拟和试验分析。结果表明,所提方法能够有效地自适应调整噪声方差和识别结构外部激励,从而实现结构响应重构。Here,aiming at the problem of the need for known external structural excitations and presetting prior constant noise variances when traditional Kalman filtering(KF)algorithm is applied in response reconstruction,a structural excitation identification and response reconstruction method based on the innovation based adaptive augmented Kalman filtering(IAAKF)algorithm was proposed.Firstly,based on the augmented state space model,external excitation vector and state vector were combined to form an augmented state vector,and according to statistical characteristics of augmented new information,Kalman filtering gain and state estimation error covariance matrix were adaptively adjusted in real-time.Secondly,using limited measured points’acceleration sensor measurement data,combining the modal method,responses of acceleration,velocity,displacement and strain at various positions of the studied structure were reconstructed.Finally,numerical simulation and test analysis were conducted for a crane truss and a simply supported beam,respectively.The results showed that the proposed method can effectively adaptively adjust noise variance and recognize structural external excitations to realize structural response reconstruction.

关 键 词：噪声方差 激励识别 新息统计 增广卡尔曼滤波 结构响应重构 

分 类 号：TH113.1[机械工程—机械设计及理论]