Physics-informed Gaussian process model for Euler-Bernoulli beam elements

<p>A physics-informed machine learning model, in the form of a multi-output Gaussian process, is formulated using Euler-Bernoulli beam equation. Given appropriate datasets, model can be used to regress analytical value structure’s bending stiffness, interpolate responses, and make probabilistic inferences on latent physical quantities. The developed applied numerically simulated cantilever beam, where regressed stiffness evaluated influence measurement noise prediction quality investigated. Further, distribution structural health monitoring context, Mahalanobis distance employed reason about possible location extent damage system. To validate framework, an experiment conducted measured heterogeneous datasets are update assumed model.</p>