A comparison study of failure prediction of composite laminate using finite element analysis and artificial neural network / Muhammad Nadiarulah Nanihar

Nanihar, Muhammad Nadiarulah (2021) A comparison study of failure prediction of composite laminate using finite element analysis and artificial neural network / Muhammad Nadiarulah Nanihar. Masters thesis, Universiti Teknologi MARA.


Composite material failure has been studied extensively for many years. However, the failure behaviour of composite materials that spontaneously fail is challenging to analyse. Hybrid composite is one of the recommended methods in modern construction to improve composites unexpected failure mode. However, because failure behaviour of hybrid composite laminates demands a lot of data, it must be predicted properly. This study's purpose is to predict the failure of hybrid composite laminates under uniaxial tension using artificial neural networks. Changing the angle of fibre orientation changed the failure behaviour of composite laminates. ANSYS Mechanical APDL 2020 was used to construct FE models to simulate physical testing. It was predicted using the ANSYS standard formulation and Maximum Stress Failure Criteria. On had already checked the model's results against acceptable public results. Using simply supported composite plates, uniaxial tension was studied. The plate has 24 layers and a layup of [0,0]. The FPF load represents the load required to attain the study's failure condition. Angle fibre orientation may affect the failure mode of composite laminates in uniaxial stress. The ANN tool in MATLAB predicted the failure. Finally, the ANSYS APDL data failure analysis was compared to the ANN model failure data. For Glass/Epoxy, Graphite/Epoxy, and Boron/Epoxy, the output failure rates are 18.97%, 6.25%, and 4.04 percent, respectively. The results of the study revealed that the method approaches produced more realistic and reliable results, with FEA results closely matching the analytical results. The current study is notable in that it advances knowledge about predicting failure behaviour in hybrid composite laminates using artificial neural networks.


Item Type: Thesis (Masters)
Email / ID Num.
Nanihar, Muhammad Nadiarulah
Email / ID Num.
Thesis advisor
Mahmud, Jamaluddin (Professor Ir. Dr.)
Subjects: Q Science > QC Physics > Composite materials
T Technology > TJ Mechanical engineering and machinery
Divisions: Universiti Teknologi MARA, Shah Alam > College of Engineering
Programme: Master of Science (Mechanical Engineering)
Keywords: Composite material; micromechanical; classical lamination theory; artificial neural network
Date: September 2021
URI: https://ir.uitm.edu.my/id/eprint/60254
Edit Item
Edit Item


[thumbnail of 60254.pdf] Text

Download (246kB)

Digital Copy

Digital (fulltext) is available at:

Physical Copy

Physical status and holdings:
Item Status:

ID Number




Statistic details