Uncertainty Prediction Output of a Finite Element Model (FEM) Using Surrogate Modelling Approach

Mohamad Syazwan Zafwan, Mohamad Suffian and Syahiir, Kamil and Ahmad Kamal, Ariffin and Abdul Hadi, Azman and Israr, M Ibrahim and Kazuhiro, Suga (2023) Uncertainty Prediction Output of a Finite Element Model (FEM) Using Surrogate Modelling Approach. In: 7TH SYMPOSIUM ON DAMAGE MECHANICS OF MATERIALS AND STRUCTURES, 19 - 20 September 2023, Bangi Resort Hotel.

PDF SDMMS ID 11.pdf Download (363kB)

Abstract

Additive Manufacturing (AM) is a manufacturing approach that can build a three-dimensional object from a computer- aided design model by adding material layer by layer. This method has gained popularity due to its capability to manufacture a product with complex geometries. However, uncertainties exist in its structure as it involves the material properties and geometry parts. A computational approach via Finite Element Method (FEM) is an alternative to overcome these uncertainties. Due to its high computational effort and time consumption, the Machine Learning approach via Surrogate Modelling is another method to produce the output of the simulation results. Surrogate Modelling can generate output with an R2 value of 0.98 intervals when compared with the FEM results. The results demonstrate the potential of Surrogate Modelling to run FEM output via sufficient training data input.

Actions (For repository members only: login required)

View Item