Computational Mechanics


The computational mechanics area has originated within traditional civil engineering areas such as structures, geomechanics, hydraulics and constructions, implicitly including and in a certain way unifying subject matters connected to analysis, simulation and modelling of physical engineering problems, with the use of computation. The area has been constantly updated, notably keeping high production levels in line with international standards and has exerted important influence in science and technology development, over the last four decades, opening the possibility of transforming classical continuum mechanic theories in effective practical instruments to predict and understand the behaviour of complex systems. The computational mechanics tools are widely seen applied to the simulation of design and current technological developments; including industry, medicine, defence and many other recognizable areas of knowledge, all without losing tract of its origins, quite present to overcome challenges and difficulties found in engineering design.

Within the several research lines of this area, it should be emphasized not only the developments but also the advanced applications of numerical methods, such as finite elements, boundary elements, finite differences, finite volumes and meshless techniques, among others, that have been developed with the help of existing computers and clusters found in the CEP. With reference to the field of applications directly connected to the area, it is worth mentioning prototypes and computational systems applied to the solution of nonlinear and time dependent problems, typical of stress analysis, fluid mechanics and potential theory in general, like: implementation of thermo-chemo-mechanical models for massive concrete structures, visco-elastoplastic material modelling, geometric nonlinear modelling, wave propagation problems, flow problems, well simulation, fire breakout simulation, cathodic protection system simulation, bioengineering, etc.

Research Lines

Discrete methods for solution of differential equations
This research line incorporates in addition to the development of classical numerical methods such as finite elements, boundary elements and finite differences, also new varieties as finite volumes and meshless techniques, among others, also including possible combinations of such methods.

High performance computing
High performance computing comprises the study and development of parallel techniques aiming to improve the use of clusters and shared or distributed memory machines. Data structure and algorithm optimization for large-scale computing are also studied in this line.

Computational algorithms and techniques
This line incorporates tools and specific algorithms aiming at improving the computational efficiency in applications of many existing numerical methods. The line includes: solution of linear and nonlinear systems of algebraic equations, numerical integration, mesh generation, eigenvalues and eigenvectors computation, time integration algorithms and further iterative procedures in general.

Geophysics – velocity model at 2650m depth
Boundary elements: 3-D simulation of electrochemical potential distribution in industrial pipes under cahodic protection
Finite elements: bio-heat diffusion modelling in human brain

Meshless methods: modelling of fracture mechanics problem

Finite elements – 3–D mesh of the powerhouse of a hydroelectric power station
Natural convection problem - Rayleigh-Benard problem solved with stabilized finite elements