Predicting impact damage in composite structures
File(s)
Author(s)
Alhashmy, Hasan Ali
Type
Thesis
Abstract
Ever since fibre composite structures were used in advanced applications, predicting damage due to low/high-velocity impact (LVI and HVI) has been a major concern during components' service life. However, current predictive modelling approaches require extensive and expensive validation testing to overcome variability and scatter in composite material input properties. Most researchers attempt to solve this issue by fitting their modelling approaches with fitting curves to match experimental results or performing numerous tests considering various variables, but these methods are invariably case-specific and rarely transferrable to different composite structures and components.
This research work aims to address these challenges by overcoming the scatter in composite material properties and damage model variabilities by performing analysis on a wider range of fibre composites. The new engineering and lifing method proposes a combination of upper/mean/lower bound of materials properties for modelling damage. Data was taken from tests and collected from a range of carbon/glass fibre composite structures at intra/inter laminar levels and used as inputs into numerical modelling.
A conservative assumption was made that surface damage due to impact is related to internal depth damage, and normalizing this measured value with net thickness issue can define a new impact damage limit. By adopting a standard micro-modelling technique with statistically determined material property data from a number of glass and carbon fibre composites strengths and damage properties from limited data sets, it is possible to determine a conservative level of safety for a range of composites subjected to impact damage.
The new method was employed in the development of a novel feature experiment to simulate impact damage in glass fibre composite ring/pipe under pressure using an external laterally loaded boundary condition applied to the surface of a pipe section. This work gives increased confidence that performing feature impact tests can reduce expensive pressurised pipe testing.
This research work aims to address these challenges by overcoming the scatter in composite material properties and damage model variabilities by performing analysis on a wider range of fibre composites. The new engineering and lifing method proposes a combination of upper/mean/lower bound of materials properties for modelling damage. Data was taken from tests and collected from a range of carbon/glass fibre composite structures at intra/inter laminar levels and used as inputs into numerical modelling.
A conservative assumption was made that surface damage due to impact is related to internal depth damage, and normalizing this measured value with net thickness issue can define a new impact damage limit. By adopting a standard micro-modelling technique with statistically determined material property data from a number of glass and carbon fibre composites strengths and damage properties from limited data sets, it is possible to determine a conservative level of safety for a range of composites subjected to impact damage.
The new method was employed in the development of a novel feature experiment to simulate impact damage in glass fibre composite ring/pipe under pressure using an external laterally loaded boundary condition applied to the surface of a pipe section. This work gives increased confidence that performing feature impact tests can reduce expensive pressurised pipe testing.
Version
Open Access
Date Issued
2023-10-14
Date Awarded
2024-06-01
Copyright Statement
Attribution-NonCommercial 4.0 International Licence (CC BY-NC)
License URL
Advisor
Blackman, Bamber
Publisher Department
Mechanical Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)