Abstract:<p>While the theory of Bayesian system identification provides a probabilistic means for reliably and robustly inferring models of a dynamic system and their parameters based on measured dynamic response, exploiting sparseness during online tracking of changing model parameters is not well understood. The focus in this study is to implement the dual Kalman filter for real-time Bayesian sequential state and parameter identification based on noisy sensor signals while incorporating sparse Bayesian learning to impose sparse model parameter changes from their initial reference values. We also want out model to be able to capture the evolution of sparse model parameter changes between two successive time instants. To this end, we present a hierarchical Bayesian model for tracking the joint posterior distribution of the state and model parameter vectors for a monitored dynamical system, where the two afore-mentioned sparseness constraints (sparse changes from reference values and with time) are also effectively incorporated for each time. We show our stochastic model of the structural dynamical system can be represented as a coupled conditionally-linear Gaussian state space model for the state and model parameter vectors, leading to some interesting analytical properties of the method that allow quantities of interest to be calculated in real time by using Kalman filtering equations. The parameters for the measurement and state prediction errors are learned solely from the available data up to the current time and so our method resolves the well-known instability problem in Kalman filtering due to arbitrary assignment of the error-distribution parameters. Finally, two illustrative applications are presented, one for identification of stiffness <a href="https://www.fema.gov/media-library-data/20130726-1719-25045-2787/fema%5Fp440a%5Fpart1.pdf">degradation</a> and the other for input time–history identification where both are based on noisy dynamic response measurements from a structure.</p>

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to effectively utilize sparsity when tracking the changes of dynamic system parameters in real - time. Specifically, the paper focuses on how to combine the sparse Bayesian learning method to impose the sparsity of model parameter changes from the initial reference value and the sparsity constraint in time when performing real - time Bayesian sequential state and parameter identification based on noisy data of sensor signals. The paper proposes a hierarchical Bayesian model for tracking the joint posterior distribution of the state and model parameter vectors of the monitored dynamic system, and effectively integrates the above two sparsity constraints at each time point. In addition, the paper also shows how to represent the stochastic model of the structural dynamic system by coupling the conditional linear Gaussian state - space model, so as to use the Kalman filter equations to obtain the quantities of interest in real - time calculations. The main contributions in the paper include: 1. **Proposing a new sparsity - inducing dual - Kalman filter**: This filter can estimate the state and model parameters simultaneously, and takes into account the change sparsity of model parameters and the sparsity in time during the estimation process. 2. **Solving the instability problem in Kalman filtering**: By using only the data at the current time and before to learn the parameters of measurement and state prediction errors, the instability caused by arbitrarily specifying the error distribution parameters is avoided. 3. **Application examples**: The paper shows the effectiveness of this method through two specific examples. One is to identify the structural stiffness degradation, and the other is to identify the time history of the input. Both of these examples are based on the noisy dynamic response measurements of the structure. In summary, the paper aims to provide an effective real - time method for tracking the sparsely changing states and parameters in dynamic systems, and has important application value especially in fields such as structural health monitoring and input identification.

Novel sparseness-inducing dual Kalman filter and its application to tracking time-varying spatially-sparse structural stiffness changes and inputs

Joint Tracking of Performance Model Parameters and System Behavior Using a Multiple-Model Kalman Filter

Online identification of time-variant structural parameters under unknown inputs basing on extended Kalman filter

Bayesian System Identification based on Hierarchical Sparse Bayesian Learning and Gibbs Sampling with Application to Structural Damage Assessment

Distributed Robust Bayesian Filtering for State Estimation.

An improved Kalman filter with dummy measurement for identification of structural load and unknown parameters

Multiple importance unscented Kalman filtering with soft spatiotemporal constraint for multi‐passive‐sensor target tracking

Multi-Fading Factor and Updated Monitoring Strategy Adaptive Kalman Filter-Based Variational Bayesian

Adaptive unscented Kalman filter methods for identifying time‐variant parameters via state covariance re‐updating

Multi-Fading Factor and Update Monitoring Strategy Adaptive Kalman Filter Based Variational Bayesian with Inaccurate Time-Varying Process and Measure-Ment Noise Covariance Matrices

Experimental validation of the Kalman-type filters for online and real-time state and input estimation

An adaptive generalized extended Kalman filter for real-time identification of structural systems, state and input based on sparse measurement

An Iterative Augmented Unscented Kalman Particle Filter for Simultaneous State-Parameter-Input Estimation for Structural Systems Subjected to Gamma-Distribution Noise

Sequential sparse Bayesian learning with applications to system identification for damage assessment and recursive reconstruction of image sequences

Latent-KalmanNet: Learned Kalman Filtering for Tracking from High-Dimensional Signals

Sparsity promoting algorithm for identification of nonlinear dynamic system based on Unscented Kalman Filter using novel selective thresholding and penalty-based model selection

Joint State and Sparse Input Estimation in Linear Dynamical Systems

Adaptive Unscented Kalman Filter for Target Tacking with Time-Varying Noise Covariance Based on Multi-Sensor Information Fusion

Joint State and Parameter Estimation for Multisensor Nonlinear Dynamic Systems on the Basis of Strong Tracking Filter

Robust adaptive Kalman filter for structural performance assessment

Synchronous identification of nonlinear structural parameters and unknown external excitation based on improved UKF