(610c) Distributed State Estimation from Delayed Measurements

Modern chemical plants often consist of highly integrated subsystems with strong material and energy interactions. In recent years, the international competition for the production of higher-quality products at lower costs has made industries pay increased attention to product quality monitoring and control [1, 2]. However, online measurements of product quality indices are often not available or available infrequently and with time delays. In such cases, frequent information on these variables can be calculated using a multirate state estimator. The combination of these measurements can improve state estimation accuracy and robustness. For large-scale plant, the distributed implementation of state estimators has been found to have several advantages [3]. Previous studies have focused on the centralized state estimation in the presence of delayed and infrequent measurements [7-9]. However, distributed state estimation in the presence of the measurement attributes has received little attention.

In this work, we propose a distributed state estimation method that utilizes sampled state augmentation approach [5, 11] to handle delays in different subsystems. A sampled state augmentation approach is use to handle the tradeoff between computational time and estimation accuracy compared to existing approaches [5]. Unscented Kalman filter [10]is chosen to test the method, as it is suitable for highly nonlinear systems. The proposed method can be extended to other nonlinear filters, such as the extended Kalman filter and moving horizon. The proposed algorithm is implemented on a process consisting of two reactors and a separator to show the improved state estimates on arrival of the delayed measurements.

REFERENCES

[1] Soroush, M., et al., Model-predictive safety: A new evolution in functional safety, in Smart Manufacturing. 2020, Elsevier. p. 283-321.

[2] Baldea, M. and P. Daoutidis, Dynamics and nonlinear control of integrated process systems. 2012: Cambridge University Press.

[3] Carli, R., et al., Distributed Kalman filtering based on consensus strategies. IEEE Journal on Selected Areas in communications, 2008. 26(4): p. 622-633.

[4] Khan, U.A. and J.M. Moura, Distributing the Kalman filter for large-scale systems. IEEE Transactions on Signal Processing, 2008. 56(10): p. 4919-4935.

[5] Gopalakrishnan, A., N.S. Kaisare, and S. Narasimhan, Incorporating delayed and infrequent measurements in Extended Kalman Filter based nonlinear state estimation. Journal of Process Control, 2011. 21(1): p. 119-129.

[6] Soroush, M., State and parameter estimations and their applications in process control. Computers & Chemical Engineering, 1998. 23(2): p. 229-245.

[7] Fatehi, A. and B. Huang, Kalman filtering approach to multi-rate information fusion in the presence of irregular sampling rate and variable measurement delay. Journal of Process Control, 2017. 53: p. 15-25.

[8] Choi, M., et al. State estimation with delayed measurements considering uncertainty of time delay. in 2009 IEEE International Conference on Robotics and Automation. 2009. IEEE.

[9] Alexander, R., et al., Challenges and Opportunities on Nonlinear State Estimation of Chemical and Biochemical Processes. Processes, 2020. 8(11): p. 1462.

[10] Wan, E.A., R. Van Der Merwe, and S. Haykin, The unscented Kalman filter. Kalman filtering and neural networks, 2001. 5(2007): p. 221-280.

[11] Van Der Merwe, R., Sigma-point Kalman filters for probabilistic inference in dynamic state-space models. 2004, OGI School of Science & Engineering at OHSU.