Gaussian Process Regression via Kalman Filtering

Author: Joanna Zou	Date: Feb. 12, 2022	Cite this page

We show that a Bayesian filtering technique which probabilistically models unknown inputs of a dynamical system can be used to efficiently estimate critical states of in-situ structures, paving a way for improved structural health monitoring and performance-based design of offshore wind turbines.

$x + y = z$

$$\begin{aligned} 2x + 3y &= 7 \\ x - y &= 1 \end{aligned}$$

References

J. Zou, E. Lourens, A. Cicirello. “Virtual sensing of subsoil strain response in monopile-based offshore wind turbines via Gaussian process latent force models.” Mechanical Systems & Signal Processing. 200 (110488). 2023.  

J. Zou, A. Cicirello, A. Iliopoulos, E. Lourens. “Gaussian process latent force models for virtual sensing in a monopile-based offshore wind turbine.” Proceedings of the European Workshop on Structural Health Monitoring, pp. 290-29. 2022.