24
2 State of the art
that it involves the combination of analysis software, methods, people skills and experience to enable structural strength predictions with high level of confidence. These and
other available resources can be broken down two three key points
Prediction of damage mechanisms and failure
Close interaction of simulations and physical tests
Hierarchical approach
The first point emphasizes that virtual testing is particularly targeted towards simulating
damage mechanisms. This requires advanced non-linear finite element solutions. In addition, deep understanding of the actual damage mechanisms and their interactions is
essential. The second point stands for the fact that physical tests are an integral part of
a virtual test framework. Tests are inevitable for validation and calibration of the simulation but also for providing the basis for understanding the damage mechanisms that
led to failure. The other way around simulation models support the test planning by
providing preliminary information about critical load cases and suitable boundary conditions. Lastly the third point indicates that all tests, simulation models and analyses are
structured hierarchically.
There are two types of hierarchy, which are common in the context of virtual testing.
The first is the multiscale analysis, where the hierarchy is according to the investigated
length scale. It can be described as the sequential coupling of different analysis models
at different scales and levels of fidelity [Ost11]. This often involves a combination of micro and macro mechanics to analyze structures in great detail [Abd09]. This can be done
for instance by determining the properties of one entity (e.g. composite ply) at a suitable
length scale, transforming the results into a constitutive model and transferring these
homogenized information to the following length scale to analyze the physical behavior
of a larger entity (e.g. laminate) [LLo13]. Typical length scales of multiscale analyses are
illustrated in Figure 20 a). This type of hierarchy is widely established in the material science context, in particular for the development of novel composite materials which can
greatly benefit from virtual testing methods [Gon07], [Oke14]. Therefore, applications
that apply multiscale analysis often end at macroscale (coupon level). The second type
of hierarchy is according to the level of structural complexity of the investigated entity.
These complexity levels are also known as ‘building blocks’, which originates from the
substantiation of aircraft components [Mil03]. Applications with building block hierarchy
usually target larger length scales (structural level), while it is not required to cover multiple length scales in a sense where constituent models are derived and integrated between length scales. However, similarly to the multiscale analysis, a fundamental part of
the building blocks is the integration of gained knowledge from one level to the other
2 State of the art
that it involves the combination of analysis software, methods, people skills and experience to enable structural strength predictions with high level of confidence. These and
other available resources can be broken down two three key points
Prediction of damage mechanisms and failure
Close interaction of simulations and physical tests
Hierarchical approach
The first point emphasizes that virtual testing is particularly targeted towards simulating
damage mechanisms. This requires advanced non-linear finite element solutions. In addition, deep understanding of the actual damage mechanisms and their interactions is
essential. The second point stands for the fact that physical tests are an integral part of
a virtual test framework. Tests are inevitable for validation and calibration of the simulation but also for providing the basis for understanding the damage mechanisms that
led to failure. The other way around simulation models support the test planning by
providing preliminary information about critical load cases and suitable boundary conditions. Lastly the third point indicates that all tests, simulation models and analyses are
structured hierarchically.
There are two types of hierarchy, which are common in the context of virtual testing.
The first is the multiscale analysis, where the hierarchy is according to the investigated
length scale. It can be described as the sequential coupling of different analysis models
at different scales and levels of fidelity [Ost11]. This often involves a combination of micro and macro mechanics to analyze structures in great detail [Abd09]. This can be done
for instance by determining the properties of one entity (e.g. composite ply) at a suitable
length scale, transforming the results into a constitutive model and transferring these
homogenized information to the following length scale to analyze the physical behavior
of a larger entity (e.g. laminate) [LLo13]. Typical length scales of multiscale analyses are
illustrated in Figure 20 a). This type of hierarchy is widely established in the material science context, in particular for the development of novel composite materials which can
greatly benefit from virtual testing methods [Gon07], [Oke14]. Therefore, applications
that apply multiscale analysis often end at macroscale (coupon level). The second type
of hierarchy is according to the level of structural complexity of the investigated entity.
These complexity levels are also known as ‘building blocks’, which originates from the
substantiation of aircraft components [Mil03]. Applications with building block hierarchy
usually target larger length scales (structural level), while it is not required to cover multiple length scales in a sense where constituent models are derived and integrated between length scales. However, similarly to the multiscale analysis, a fundamental part of
the building blocks is the integration of gained knowledge from one level to the other