Repositório Colecção:http://hdl.handle.net/10198/4202014-07-10T13:06:03Z2014-07-10T13:06:03ZMultilocal programming: a derivative-free filter multistart algorithmFernandes, Florbela P.Costa, M. Fernanda P.Fernandes, Edite M.G.P.http://hdl.handle.net/10198/90262014-04-17T01:13:27Z2013-01-01T00:00:00ZTítulo: Multilocal programming: a derivative-free filter multistart algorithm
Autor: Fernandes, Florbela P.; Costa, M. Fernanda P.; Fernandes, Edite M.G.P.
Resumo: Multilocal programming aims to locate all the local solutions
of an optimization problem. A stochastic method based on a multistart
strategy and a derivative-free filter local search for solving general constrained
optimization problems is presented. The filter methodology is
integrated into a coordinate search paradigm in order to generate a set
of trial approximations that might be acceptable if they improve the
constraint violation or the objective function value relative to the current
one. Preliminary numerical experiments with a benchmark set of
problems show the effectiveness of the proposed method.2013-01-01T00:00:00ZA derivative-free filter driven multistart technique for global optimizationFernandes, Florbela P.Costa, M. Fernanda P.Fernandes, Edite M.G.P.http://hdl.handle.net/10198/86212013-09-24T09:54:14Z2012-01-01T00:00:00ZTítulo: A derivative-free filter driven multistart technique for global optimization
Autor: Fernandes, Florbela P.; Costa, M. Fernanda P.; Fernandes, Edite M.G.P.
Resumo: A stochastic global optimization method based on a multistart strategy and a derivative-free filter local search for general constrained optimization is presented and analyzed. In the local search procedure, approximate descent directions for the constraint violation or the objective function are used to progress towards the optimal solution. The algorithm is able to locate all the local minima, and consequently, the global minimum of a multi-modal objective function. The performance of the multistart method is analyzed with a set of benchmark problems and a comparison is made with other methods.2012-01-01T00:00:00ZMultilocal programming and applicationsPereira, Ana I.Ferreira, OlgaPinho, SimãoFernandes, Edite M.G.P.http://hdl.handle.net/10198/82892014-04-17T01:10:15Z2013-01-01T00:00:00ZTítulo: Multilocal programming and applications
Autor: Pereira, Ana I.; Ferreira, Olga; Pinho, Simão; Fernandes, Edite M.G.P.
Resumo: Multilocal programming aims to identify all local maximizers of unconstrained or constrained nonlinear optimization problems. The multilocal programming theory relies on global optimization strategies combined with simple ideas that are inspired in deflection or stretching techniques to avoid convergence to the already detected local maximizers. The most used methods to solve this type of problems are based on stochastic procedures. In general, population-based methods are computationally expensive but rather reliable in identifying all local solutions. Stochastic methods based on point-to-point strategies are faster to identify the global solution, but sometimes are not able to identify all the optimal solutions of the problem.
To handle the constraints of the problem, some penalty strategies are proposed.
A well-known set of test problems is used to assess the performance of the algorithms.
In this chapter, a review on recent techniques for both unconstrained and
constrained multilocal programming is presented. Some real-world multilocal programming problems based on chemical engineering process design applications are
described.2013-01-01T00:00:00ZGranular beds composed of different particle sizes: experimental and CFD approachesDias, Ricardo P.Fernandes, Carla S.http://hdl.handle.net/10198/49232013-06-04T10:25:38Z2009-01-01T00:00:00ZTítulo: Granular beds composed of different particle sizes: experimental and CFD approaches
Autor: Dias, Ricardo P.; Fernandes, Carla S.
Resumo: The porosity, tortuosity, permeability and heat exchange characteristics from binary
packings, containing mixtures of small d and large D spherical particles, are analysed in
the present work. Binary packing porosity (ε), tortuosity (τ), permeability and heat
exchange performance are dependent on the volume fraction of large particles, xD,
present in the mixtures, as well as on the particle size ratio, δ = d/D. In the region of
minimum porosity from the binary mixtures (containing spheres with diameter d and D),
heat exchange performance and permeability from binary packing are higher than that of
the packing containing the small particles d alone (mono-size packing). The δ region
where the permeability of binary packing is higher than the permeability of mono-size
packing of particles d is located in the range 0.1 ≤ δ < 1.0. An increase in permeability
by a factor of two is achieved for particle size ratios between 0.3 and 0.5. Tortuosity can
be modelled by the simple function τ = 1/ε α and it is shown that, in the region of minimum porosity, α varies between 0.5 (mono-size packing) and 0.4 (binary packing
with δ close to 0.03). Due to the tortuosity increase, binary mixtures give rise to
Kozeny´s coefficients substantially higher than five.
Using the commercial finite element software package POLYFLOW® it was possible
to confirm the heat exchange enhancement referred above. The obtained improvement on
the thermal performance is related to the increase of effective thermal conductivity in the
binary packing and to the increase in transversal thermal conductivity due to the porosity
decrease and tortuosity increase. For non-Newtonian fluids from the power-law type, τ decreases with the decrease of the flow index behaviour2009-01-01T00:00:00Z