7 March 2022 in IEEE Transactions on Industrial Electronics
Sensors provide insights into the industrial processes, while misleading sensor outputs may result in inappropriate decisions or even catastrophic accidents. In this article, the Bayesian estimation algorithms are developed to estimate unforeseen signals in sensor outputs without tuning. The optimal Bayesian estimation method is first derived by incorporating a Gaussian distribution specifying potential unmodeled dynamics into the measurement equation. Since its performance depends on tuning parameters, an iterative Bayesian estimation algorithm is developed using the variational inference technique. Specifically, an inverse Wishart distribution is introduced to describe the predicted covariance of abnormal signals. We then estimate it together with the other independent Gaussian distributions to conditionally approximate the joint posterior distribution, by which the effects of tuning parameters can be replaced adaptively. Testing the proposed algorithms through a simulated electromechanical brake model and a real experimental system shows that the proposed algorithm can satisfactorily estimate additive sensor faults online and services as a sensor monitor that simultaneously provides the locations and magnitudes of faulty signals without tuning.
1 April 2021 in International Journal of Adaptive Control and Signal Processing
15 April 2024 in IEEE/CAA Journal of Automatica Sinica
In the era of exponential growth of data availability, the architecture of systems has a trend toward high dimensionality, and directly exploiting holistic information for state inference is not always computationally affordable. This paper proposes a novel Bayesian filtering algorithm that considers algorithmic computational cost and estimation accuracy for high-dimensional linear systems. The high-dimensional state vector is divided into several blocks to save computation resources by avoiding the calculation of error covariance with immense dimensions. After that, two sequential states are estimated simultaneously by introducing an auxiliary variable in the new probability space, mitigating the performance degradation caused by state segmentation. Moreover, the computational cost and error covariance of the proposed algorithm are analyzed analytically to show its distinct features compared with several existing methods. Simulation results illustrate that the proposed Bayesian filtering can maintain a higher estimation accuracy with reasonable computational cost when applied to high-dimensional linear systems.
23 July 2024 in Computers & Chemical Engineering
27 January 2026 in IEEE Transactions on Industrial Informatics
Effective state estimation is crucial for ensuring operational safety, environmental compliance, and efficient resource utilization in high-dimensional wastewater treatment plants (WWTPs). This study proposes a purely data-driven methodology based on dynamic mode decomposition (DMD) to construct a linear dynamic model capable of globally capturing the inherent nonlinearity and complexity of wastewater treatment processes. Moreover, DMD can generate system representations in both full-dimension and reduced-dimension forms and offers flexibility in modeling complex dynamics. Then, the Kalman filter and ensemble Kalman filter (EnKF) are implemented using the derived models for state estimation. A detailed analysis of estimation accuracy and computational cost is also provided. Furthermore, the performance of the proposed DMD-based method is comparatively evaluated against a traditional linearization-based approach, in which the process is linearized at a nominal operating point. Results from the high-dimensional WWTP demonstrate the effectiveness and superiority of the proposed method.
6 May 2026 in IEEE Transactions on Instrumentation and Measurement
This paper considers the problem of parameter identification for linear parameter-varying (LPV) systems with asymmetric noise. Commonly employed noise distributions exhibit symmetric statistical properties that make it difficult to describe asymmetric measurement noise. This leads to a degradation in the performance of the identification algorithm. To cope with the asymmetric noise, the measurement noise is characterized by the generalized hyperbolic skew Student’s t (GHS) distribution. Then, the robust variational algorithm for the LPV system is proposed within the Bayesian framework. The variational Bayesian (VB) algorithm estimates the parameters of the model and the asymmetric noise from the measured data. The proposed algorithm is validated through a numerical example and a mass–spring–damper system. The results and findings show that the proposed algorithm can effectively address the problem caused by asymmetric measurement noise. It can accurately identify the model parameters and suppress the influence of the asymmetric noise.