Machine learning aided stochastic free vibration analysis of functionally graded porous plates

A machine learning aided uncertainty quantification framework is introduced for the free vibration analysis (MLA-SFV) of functionally graded porous structures. The proposed framework solves the uncertain free vibration problems with stochastic eigenvalue results. Various types of systematic properties can be modeled as random variables into the formulation level. To effectively solve the stochastic free vibration questions, two popular machine learning techniques, the extended support vector regression, along with a typical algorithm named the Gaussian process are introduced. Through the MLA-SFV framework, the statistical moments, the probability density function, and the cumulative density function of random eigenvalues can be acquired in an efficient manner. Detailed numerical investigations have been implemented to illustrate the accuracy, efficiency, and applicability of the MLA-SFV framework for functionally graded porous structures.