SULEIMAN M. S. BANI HANI

Research Interests
Artificial Intelligence 
Applying genetic algorithms for numerical integration on complex domains for meshfree methods.
Applying fuzzy logic and neural networks for nonlinear control problems.
Using genetic algorithms for optimal control problems.
Numerical integration in R3:
In this thesis we discuss efficient techniques for numerical integration in R1
and R2. The next step is to expand numerical integration for the method of
finite spheres to three-dimensions. Three-dimensional numerical integration
on an arbitrary domain is the first step in applying the method of finite
spheres to more realistic and complex problems.
• Application to large deformation and fracture problems:
One of the key advantages to the method of finite spheres is the potential to
model problems with large deformation and fracture without the need for
remeshing. Our next step in this research is to advance the computational
technology to handle such problems. The use of the lookup table approach
for the total Lagrangian technique is straightforward. However, its use with
the updated Lagrangian formulation would require further work.
• Application to multiscale problems:
Although there is significant research to develop continuum theories for
problems with micron and nano scale, most of these theories cannot cover
the wide range of material behavior at the different length scales. Multiscale
modeling is being widely investigated because of its ability to describe and
model material behavior at deferent length scales, and using meshfree
methods such as the method of finite spheres offer a great flexibility for
adopting and incorporating different models and material behavior. There are
two ways the new integration method may be used to advantage. In
hierarchical multiscale methods such as the ones based on asymptotic
homogenization and structural enrichment-based techniques, highly complex
domain integrals arise which may be efficiently computed using this method.
In concurrent multiscale methods, where continuum and atomistic methods
are coupled in a single simulation scheme, the new integration approach may
be used for the continuum for seamless message passing to and from the
nonlocal parts of the computational domain.
• Application to other meshfree methods:
Most meshfree methods use rational nonpolynomial interpolation functions
and the integration domains are usually more complex the in the traditional
finite element method. Using the new numerical integration method with
different meshfree methods will result in a more efficient approximation
schemes and will open the opportunity to explore the potential of these
methods.
SULEIMAN M. S. BANI HANI

SULEIMAN M. S. BANI HANI

Welcome to the Hashemite University faculty staff website.
